Taming numbers and durations in the model checking integrated planning system
- 格式:pdf
- 大小:649.91 KB
- 文档页数:44
CHAPTER 1313-1. Suppose there are 25,000 unemployed persons in the economy. You are given the following data about the length of unemployment spells:Duration of Spell(in months) ExitRate1 0.602 0.203 0.204 0.205 0.206 1.00where the exit rate for month t gives the fraction of unemployed persons who have been unemployed t months and who “escape” unemployment at the end of the month.The data can be used in the problem to calculate the number of workers who have 1 month of unemployment, the number who have 2 months of unemployment, and so on, and how many months of unemployment are associated with workers who get a job after a given duration.Duration (Months) ExitRate# Unemp:Start ofMonth# ofExiters# ofStayers# MonthsForDuration1 0.60 25,000 15,00010,000 15,0002 0.20 10,000 2,000 8,000 4,0003 0.20 8,000 1,600 6,400 4,8004 0.20 6,400 1,280 5,120 5,1205 0.20 5,120 1,024 4,096 5,1206 1.00 4,096 4,096 0 24,576(a) How many unemployment-months will the 25,000 unemployed workers experience?The 25,000 workers will experience 58,616 months of unemployment, an average of 2.34 months per worker.(b) What fraction of persons who are unemployed are “long-term unemployed” (that is, are in unemployment spells which last 5 or more months)?Only 5,120 of the 25,000 workers (20.5 percent) are in spells lasting 5 or more months.(c) What fraction of unemployment months can be attributed to persons who are long-term unemployed?Although only 20.5 percent of workers are unemployed for 5 or more months, they account for 29,696 of the 58,616 (50.7 percent) of months of unemployment.(d) What is the nature of the unemployment problem in this example: too many workers losing their jobs or too many long spells?Most spells are short-lived, but workers in long spells account for most of the unemployment observed in this economy. Thus, the main problem is too many long spells.13-2. Consider Table 599 of the 2002 U.S. Statistical Abstract.(a) How many workers aged 20 or older were unemployed in the United States during 2001? How many of these were unemployed less than 5 weeks, 5 to 14 weeks, 15 to 26 weeks, and 27 or more weeks?In total, 5,554,000 workers aged 20 years or older were unemployed in the U.S. in 2001. Of these, 40 percent, or 2,221,600 workers, were unemployed for less than 5 weeks; 32.2 percent, or 1,788,388 workers, were unemployed between 5 and 14 weeks; 15.5 percent, or 860,870 workers, were unemployed15 to 26 weeks, and 12.9 percent, or 716,466 workers were unemployed for 27 or more weeks.(b) Assume that the average spell of unemployment is 2.5 weeks for anyone unemployed for less than 5 weeks. Similarly, assume the average spell is 10 weeks, 20 weeks, and 35 weeks for the remaining categories. How many weeks did the average unemployed worker remain unemployed? What percent of total months of unemployment are attributable to the workers that remained unemployed for at least 15 weeks?The total number of weeks of unemployment is calculated as:2,221,600(2.5) + 1,788,388(10) + 860,870(20) + 716,466(35) = 65,731,590months of unemployment spread over 5,554,000 unemployed workers, implies that the average unemployed worker remained unemployed for 11.835 weeks. The percent of total months of unemployment attributable to the workers that remained unemployed for at least 15 weeks is[ 860,870(20) + 716,466(35) ] / 65,731,590 = 64.34 percent.13-3. Suppose the marginal revenue from search isMR = 50 - 1.5w,where w is the wage offer at hand. The marginal cost of search isMC = 5 + w.(a) Why is the marginal revenue from search a negative function of the wage offer at hand?If the offer-at-hand is relatively low, it pays to keep on searching as the next offer is likely higher than the offer-at-hand. If the offer-at-hand is very high, however, it does not pay to keep on searching since it is unlikely that the next search will generate a higher wage offer.(b) Can you give an economic interpretation of the intercept in the marginal cost equation; in other words, what does it mean to say that the intercept equals $5? Similarly, what does it mean to say that the slope in the marginal cost equation equals one dollar?The $5 indicates the out-of-pocket search costs. Even if the offer-at-hand is zero (so that there is no opportunity cost to search), it still costs money to get to the firm and learn about the details of the potential job offer. The slope equals $1, because the costs of search also vary directly with the opportunity cost of search which is the wage offer at hand. If the wage offer at hand is $10, the opportunity cost from one more search equal $10; if the wage offer at hand is $11, the opportunity cost would be $11, and so on.(c) What is the worker’s asking wage? Will a worker accept a job offer of $15?The asking wage is obtained by equating the marginal revenue of search to the marginal cost of search, or 50 – 1.5w = 5 + w. Solving for w implies that the asking wage is $18. The worker, therefore, would not accept a job offer of $15.(d) Suppose UI benefits are reduced, causing the marginal cost of search to increase to MC = 20 + w. What is the new asking wage? Will the worker accept a job offer of $15?If we equate the new marginal cost equation to the marginal revenue equation we find that the asking wage drops to $12. The worker will now accept a wage offer of $15.13-4. (a) How does the exclusion of non-working welfare recipients affect the calculation of the unemployment rate? Use the 2002 U.S. Statistical Abstract to estimate what the 2000 unemployment rate would have been if welfare recipients had been included in the calculation.Excluding non-working welfare recipients from the unemployment rate biases the unemployment rate downward. Table 560 of the Statistical Abstract shows that the labor force in 2000 totaled 140,863,000, while the number employed totaled 135,208,000. Thus, the unemployment rate was 4.0 percent.Table 513 of the Statistical Abstract shows that 2,253,000 people received public assistance in 2000. Assuming all of these were non-working and out of the labor force, their inclusion in the calculation of the unemployment rate would increase the unemployment rate to(140,863,000 + 2,253,000 – 135,208,000) / (140,863,000 + 2,253,000) = 5.5%.(b) How does the exclusion of black market workers affect the calculation of the unemployment rate? Estimate, the best you can, what the 2000 unemployment rate would have been if workers in the underground economy had been included in the calculation.Excluding black market workers from the calculation of the unemployment rate keeps the unemployment rate artificially high. At the best, black market workers are not counted in the labor force, but at the worst, the black market workers claim to be in the labor force and without a job. The problem is that there is no (or very little) data on the black market by definition. Some researchers have estimated the underground economy to be on the order of 10 to 20 percent of activity in the U.S. Suppose half of underground economy workers have regular market jobs as well. Suppose further that of the remaining half, one half claim to be unemployed while the other half is out of the labor force. Thus, 10 percent of the unemployed workers are the number of underground economy workers in the labor force without a job: 10 percent of(140,863,000 – 135,208,000) = 565,500. An equal number say they are not in the labor force. (And twice this number is already reporting having a legitimate job.) Using these estimates, therefore, the labor force should be 140,863,000 + 565,500 = 141,428,500, while the number employed should be 135,208,000 + 565,500 + 565,500 = 136,339,000. Thus, the unemployment rate would be(141,428,500 – 136,339,000) / 141,428,500 = 3.6%.13-5. Compare two unemployed workers; the first is 25 years old and the second is 55 years old. Both workers have similar skills and face the same wage offer distribution. Suppose that both workers also incur similar search costs. Which worker will have a higher asking wage? Why? Can search theory explain why the unemployment rate of young workers differs from that of older workers?The marginal revenue of search depends on the length of the payoff period. Younger workers have the most to gain from obtaining higher paying jobs, since they can then collect the returns from their search investment over a longer expected work-life. As a result, it pays for younger workers to set their asking wage at a relatively high level. This implies that younger workers will tend to have higher unemployment rates and longer spells of unemployment than older workers.13-6. Suppose the government proposes to increase the level of UI benefits for unemployed workers.A particular industry is now paying efficiency wages to its workers in order to discourage them from shirking. What is the effect of the proposed legislation on the wage and on the unemployment rate for workers in that industry?The introduction of UI benefits shifts the no-shirking supply curve upwards (from NS to NS′), because a higher wage would have to be paid in order to attract the same number of workers who do not shirk. As a result, the new equilibrium (point Q′) entails a higher efficiency wage and leads to a larger number of unemployed workers.Dploym ent13-7. It is well known that more-educated workers are less likely to be unemployed and haveshorter unemployment spells than less-educated workers. Which theory(ies), the job search model, the sectoral shifts hypothesis, or the efficiency wage model, can explain this empirical correlation?Two of the theories in question can be formulated in such a way that they each predict that more-educated workers will have less unemployment and shorter spells of unemployment than less-educated workers.The job search model suggests that highly educated workers would have a lower unemployment rate either if they have relatively higher search costs or relatively lower gains from search. It could then be argued, for example, that search costs are (relatively) higher for highly-educated workers than for less-educated workers, perhaps because the nature of the job match between a highly educated worker and a firm is much more complex than the type of job match required between the firm and a worker who does repetitive tasks.Similarly, it is plausible to argue that it is easier to retool a highly-educated worker than a less-educated worker. The decline in demand in particular industries, then, leads to sectoral shift that can be better weathered by highly-educated workers.The efficiency wage model, however, has a harder time explaining the correlation. Presumably, the output of highly-educated workers is more difficult to measure than the output of less-educated workers. As a result, efficiency wages are more likely to arise in industries that employ highly-educated workers, and these industries (which pay above-market wages) would have to maintain a pool of unemployed workers in order to keep the employed workers in line.13-8. Suppose a country has 100 million inhabitants. The population can be divided into the employed, the unemployed, and the persons who are out of the labor force (OLF). In any given year, the transition probabilities among the various categories are given by:Moving Into:Employed Unemployed OLFEmployed 0.94 0.02 0.04 Unemployed 0.20 0.65 0.15 Moving From: OLF 0.05 0.03 0.92These transition probabilities are interpreted as follows. In any given year, 2 percent of the workers who are employed become unemployed; 20 percent of the workers who are unemployed find jobs, and so on. What will be the steady-state unemployment rate?Use E for the number of employed people, U for the number of unemployed, and N for the number not participating. The flows of people among the three categories can be found by multiplying these numbers with respective probabilities in the table. In the steady-state, the flows into each category must exactly balance the outflows. This produces the following equations:Employed: E = .94E + .20U + .05NUnemployed: U = .02E + .65U + .03NOLF: N = .04E + .15U + .92NOnly two of the three equations can be used, however, as E + U + N = 100 million. The steady-state solutions for E, U, and N can now be found with brute force algebra. The solution is that 54.273 million are employed, 6.466 million are unemployed, and 39.261 million are not in the labor force. The steady-state unemployment rate is then%646.10739.60466.6100100=×=×+=E U U u .13-9. Consider an economy with 250,000 adults, of which, 40,000 are retired senior citizens, 20,000 are college students, 120,000 are employed, 8,000 are looking for work, and 62,000 stay at home. What is the labor force participation rate? What is the unemployment rate?The labor force participation rate = ( 120,000 + 8,000 ) / 250,000 = 51.2 percent.The unemployment rate = 8,000 / ( 120,000 + 8,000 ) = 6.25 percent.13-10. Consider an economy with 3 types of jobs. The table below shows the jobs, the frequency with which vacancies open up on a yearly basis, and the income associated with each job. Searching for a job costs $C per year and generates at most 1 job offer. There is a 20 percent chance of not generating any offer in a year. (Note: the expected search duration for a job with probability p of appearing is 1/p years.)Job Type Frequency IncomeA 30 percent $60,000B 20 percent $100,000C 30 percent $80,000As a function of C , specify the optimal job search strategy if the worker maximizes her expected income net of search costs.There are four possible strategies for the worker: accept any job paying at least $100,000, accept any job paying at least $80,000, accept any job paying at least $60,000, or do not search for a job.Under the first strategy, the worker will only accept a B job. The expected search time for a B job is 5 years. Thus, the expected value of this strategy is $100,000 – 5C .Under the second strategy, the worker will accept a B or a C job. The expected search time, therefore, is 2 years (i.e., there is a 50 percent chance of receiving a B or a C job in a year, so the search time is 1/0.5 =2). Thus, the expected value of this strategy isC C 2000,88$2)000,80($3.02.03.0)000,100($3.02.02.0−=−+++.Under the third strategy, the worker will accept any of the three jobs. The expected search time, therefore, is 1.25 years, and the expected value of this strategy isC C 25.1500,77$25.1)000,60($8.03.0)000,80($8.03.0)000,100($8.02.0−=−++.Under the fourth strategy, the worker does not search, and therefore does not find a job or incur search costs, for an expected payoff of $0.One can now compare the four expected payoffs to determine that the optimal search strategy is:Accept the first B job if C < $4,000.Accept the first B or C job if $4,000 # C < $14,000.Accept the first job of any kind if $14,000 # C < $62,000.Don’t search if $62,000 # C .13-11. (a) A country is debating whether to fund a national database of job openings and giving all unemployed workers free access to it. What effect would this plan have on the long-rununemployment rate? What effect would this plan have on the average duration of unemployment? Why?Lower search costs would probably increase the unemployment rate. Lower search costs (similar to extended UI benefits) will cause workers to keep their asking wage higher for a longer time. Thus, unemployed workers will be more selective in the jobs they accept. The result is that workers remain unemployed for longer periods of time. It is also possible that lower search costs will entice workers to quit their jobs in order to look for a better job. This will also increase the unemployment rate.(b) A country is debating whether to impose a $10,000 tax on employers for every worker they lay-off. What effect would this plan have on the long-run unemployment rate? What effect would this plan have on the average duration of unemployment?This policy would cause firms, in the long-run, to be very cautious in hiring new workers. (This cautious behavior is evident in Japan and some countries in Europe. The UI system in the US may also cause firms to be a bit cautious in hiring new workers.) Thus, the unemployed will remain unemployed for longer durations. The effect on the unemployment rate is a little ambiguous. The unemployment rate wouldlikely increase in the long-run, but one could also argue that the end result will be less labor turnover and, therefore, a lower rate of unemployment, especially if unemployed workers under the proposed rules are more easily discouraged and exit the labor force all together.。
高二英语经济预测单选题50题1. The GDP of a country measures the total value of all final goods and services produced within a country's borders _.A. in a given yearB. for several yearsC. since its establishmentD. in a future year答案:A。
解析:GDP(国内生产总值)是衡量一个国家在特定的一年里,在其境内生产的所有最终商品和服务的总价值。
选项A“in a given year” 在给定的一年)符合GDP的定义中关于时间的界定。
选项B“for several years” 好几年)不是GDP计算的常规时间跨度。
选项C“since its establishment”(自从它建立以来)这种时间界定不符合GDP的计算方式。
选项D“in a future year”(在未来的一年),GDP 是对已经发生的生产成果的衡量,不是未来的。
2. Inflation refers to _.A. a decrease in the general price levelB. an increase in the general price levelC. a stable price levelD. a random change in price level答案:B。
解析:通货膨胀(Inflation)指的是总体物价水平的上升。
选项A“a decrease in the general price level” 总体物价水平下降)是通货紧缩的概念。
选项C“a stable price level”( 稳定的物价水平)与通货膨胀概念相悖。
选项D“a random change in price level”(物价水平随机变化)没有准确表达通货膨胀是物价上升这一概念。
阅读是工作和日常生活中获取信息的一项重要手段,也是语言获得的重要途径之一。
近年来,区域经济的发展带动了对商务英语阅读教学的关注。
本文以高职学生为研究对象,以能力为导向,在分析商务英语阅读中图式的特点和功能的基础上,探讨图式理论对提高商务英语专业学生阅读能力的促进作用。
一、图式的概念和分类图式是由英国心理学家巴特利特(Bartlett,F.)提出,图式是阅读者对世界知识的总和,是读者从不同经历中获取经验和知识,并按情景分门别类储存于大脑的知识结构的立体网络。
20世纪70年代后期,美国人工智能专家鲁美哈特在前人研究的基础上,将图式的概念发展成为一种系统、完整的理论——图式理论。
该理论指出:当读者在理解新事物时,需要将事物与已知的观念、过去的经历和已有知识联系起来。
Cook(1989)认为图式是指读者头脑中的“先存知识”和“背景知识”。
图式在阅读理解过程中起着重要作用。
根据Carrell and Eisterhold(1983)的图式理论,图式可分为语言、形式和内容三类。
读者理解语言材料的最有效的过程就是头脑中的图式与语言材料所提供的信息之间相互作用的过程。
在此过程中,读者根据已有的背景知识和经验在“自上而下”的过程中对阅读材料做出推测;在“自下而上”的过程中对文字、语法等进行解码,构成篇章意义。
在商务英语阅读中,三类图式具备了一定的特点,在交换的阅读过程中共同作用已达到有效的理解。
语言图式的特点主要体现在商务词汇的商务背景、一词多义和大量的术语及缩写词上;形式图式的特点则由商务阅读文本约定俗成的结构和格式来体现;内容图式的特点则表现为商务学科特点的内容和文化背景。
要达到顺利而有效的商务阅读理解,读者就必须具备一定的图式,并在交互阅读过程中有效地激活相关图式。
图式的缺失或未能激活都将导致阅读理解的失败。
二、语言图式“自下而上”(the bottom-up process )这种阅读方式的采用大多由语言图式产生。
Financial Markets and Institutions, 8e (Mishkin)Chapter 20 The Mutual Fund Industry20.1 Multiple Choice1) At the beginning of 2013, mutual funds held about ________ of the U.S. stock market was held by mutual funds.A) 30%B) 50%C) 10%D) 70%Answer: ATopic: Chapter 20.1 The Growth of Mutual FundsQuestion Status: Updated from Previous Edition2) The origins of mutual funds can be traced back to the mid to late 1800s in________.A) England and ScotlandB) New York CityC) BostonD) GermanyAnswer: ATopic: Chapter 20.1 The Growth of Mutual FundsQuestion Status: New Question3) ________ intermediation means that small investors can pool their funds with other investors to purchase high face value securities.A) LiquidityB) FinancialC) DenominationD) ShareAnswer: CTopic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Previous Edition4) Mutual funds offer investors all of the following exceptA) greater-than-average returns.B) diversified portfolios.C) lower transaction costs.D) professional investment management.Answer: ATopic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Previous Edition5) Mutual fundsA) pool the resources of many small investors by selling these investors shares and using the proceeds to buy securities.B) allow small investors to obtain the benefits of lower transaction costs in purchasing securities.C) provide small investors a diversified portfolio that reduces risk.D) do all of the above.E) do only A and B of the above.Answer: DTopic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Previous Edition6) ________ enables mutual funds to consistently outperform a randomly selected group of stocks.A) Managerial expertiseB) DiversificationC) Denomination intermediationD) None of the aboveAnswer: DTopic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Previous Edition7) At the end of 2012 there were over ________ separate mutual funds with total assets over ________.A) 800; $10 trillionB) 7,500; $13 trillionC) 10,000; $10 trillionD) 1,000; $7 trillionAnswer: BTopic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Updated from Previous Edition8) Most mutual funds are structured in two ways. The most common structure is a(n) ________ fund, from which shares can be redeemed at any time at a price that is tied to the asset value of the fund. A(n) ________ fund has a fixed number of nonredeemable shares that are traded in the over-the-counter market.A) closed-end; open-endB) open-end; closed-endC) no-load; closed-endD) no-load; loadE) load; no-loadAnswer: BTopic: Chapter 20.3 Mutual Fund StructureQuestion Status: Previous Edition9) Which of the following is an advantage to investors of an open-end mutual fund?A) Once all the shares have been sold, the investor does not have to put in more money.B) The investors can sell their shares in the over-the-counter market with low transaction fees.C) The fund agrees to redeem shares at any time.D) The market value of the fund's shares may be higher than the value of the assets held by the fund.Answer: CTopic: Chapter 20.3 Mutual Fund StructureQuestion Status: Previous Edition10) The net asset value of a mutual fund isA) determined by subtracting the fund's liabilities from its assets and dividing by the number of shares outstanding.B) determined by calculating the net price of the assets owned by the fund.C) calculated every 15 minutes and used for transactions occurring during the next 15-minute interval.D) calculated as the difference between the fund's assets and its liabilities. Answer: ATopic: Chapter 20.3 Mutual Fund StructureQuestion Status: Previous Edition11) ________ funds are the simplest type of investment funds to manage.A) BalancedB) Global equityC) GrowthD) IndexAnswer: DTopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition12) The majority of mutual fund assets are now owned byA) individual investors.B) institutional investors.C) fiduciaries.D) business organizations.E) retirees.Answer: ATopic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Previous Edition13) Capital appreciation funds select stocks of ________ and tend to be ________ risky than total return funds.A) large, established companies that pay dividends regularly; moreB) large, established companies that pay dividends regularly; lessC) companies expected to grow rapidly; moreD) companies expected to grow rapidly; lessAnswer: CTopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition14) From largest to smallest in terms of total assets, the four classes of mutual funds areA) equity funds, bond funds, hybrid funds, money market funds.B) equity funds, money market funds, bond funds, hybrid funds.C) money market funds, equity funds, hybrid funds, bond funds.D) bond funds, money market funds, equity funds, hybrid funds.Answer: BTopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition15) Measured by assets, the most popular type of bond fund is the ________ bond fund.A) state municipalB) strategic incomeC) governmentD) high-yieldAnswer: BTopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition16) People who take their money out of insured bank deposits to invest in uninsured money market mutual funds have ________ risk because money market funds invest in ________ assets.A) high; long-termB) low; short-termC) high; short-termD) low; long-termAnswer: BTopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition17) The largest share of assets held by money market mutual funds isA) Treasury bills.B) certificates of deposit.C) commercial paper.D) repurchase agreements.Answer: CTopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition18) Which of the following is a feature of index funds?A) They have lower fees.B) They select and hold stocks to match the performance of a stock index.C) They do not require managers to select stocks and decide when to buy and sell.D) All of the above.Answer: DTopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition19) A deferred-load mutual fund charges a commissionA) when shares are purchased.B) when shares are sold.C) both when shares are purchased and when they are sold.D) when shares are redeemed.Answer: DTopic: Chapter 20.5 Fee Structure of Investment FundsQuestion Status: Previous Edition20) Over the past twenty years, mutual fund fees have ________, largely because________.A) fallen; SEC fee disclosure rules have led to greater competitionB) risen; investors have learned that funds with high fees provide better performanceC) risen; there has been collusion between large mutual fund companiesD) fallen; advances in information technology have lowered transaction costs Answer: ATopic: Chapter 20.5 Fee Structure of Investment FundsQuestion Status: Previous Edition21) Which of the following is most likely to be a no-load fund?A) Value fundsB) Hedge fundsC) Growth fundsD) Index fundsAnswer: DTopic: Chapter 20.5 Fee Structure of Investment FundsQuestion Status: Previous Edition22) When investors switch between funds within the same fund family, mutual funds may chargeA) a contingent deferred sales charge.B) a redemption fee.C) an exchange fee.D) 12b-1 fees.E) an account maintenance fee.Answer: CTopic: Chapter 20.5 Fee Structure of Investment FundsQuestion Status: Previous Edition23) The Securities Acts of 1933 and 1934 did notA) regulate the activities of investment funds.B) require funds to register with the SEC.C) include antifraud rules covering the purchase and sale of fund shares.D) apply to investment funds.Answer: BTopic: Chapter 20.6 Regulation of Mutual FundsQuestion Status: Previous Edition24) The largest share of total investment in mutual funds is inA) stock funds.B) hybrid funds.C) bond funds.D) money market funds.Answer: ATopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition25) Over ________ of the total daily volume in stocks is due to institutions initiating trades.A) 70%B) 50%C) 25%D) 90%Answer: ATopic: Chapter 20.6 Regulation of Mutual FundsQuestion Status: New Question26) Hedge funds areA) low risk because they are market-neutral.B) low risk if they buy Treasury bonds.C) low risk because they hedge their investments.D) high risk because they are market-neutral.E) high risk, even though they may be market-neutral.Answer: ETopic: Chapter 20.7 Hedge FundsQuestion Status: Previous Edition27) The near collapse of Long Term Capital Management was caused byA) the high management fees charged by the fund's two Nobel Prize winners.B) the fund's high leverage ratio of 20 to 1.C) a sharp decrease in the spread between corporate bonds and Treasury bonds.D) a sharp increase in the spread between corporate bonds and Treasury bonds.E) the fund's shift away from a market-neutral investment strategy.Answer: DTopic: Chapter 20.7 Hedge FundsQuestion Status: Previous Edition28) Conflicts arise in the mutual funds industry because ________ cannot effectively monitor ________.A) investment advisers; directorsB) directors; shareholdersC) shareholders; investment advisersD) investment advisers; stocks that will outperform the overall marketAnswer: CTopic: Chapter 20.8 Conflicts of Interest in the Mutual Fund IndustryQuestion Status: Previous Edition29) Late trading is the practice of allowing orders received ________ to trade at the ________ net asset value.A) before 4:00 PM; 4:00 PMB) after 4:00 PM; 4:00 PMC) after 4:00 PM; next day'sD) before 4:00 PM; previous day'sAnswer: BTopic: Chapter 20.8 Conflicts of Interest in the Mutual Fund IndustryQuestion Status: Previous Edition30) Market timingA) takes advantage of time differences between the east and west coasts of the United States.B) takes advantage of arbitrage opportunities in foreign stocks.C) takes advantage of the time lag between the receipt and execution of orders.D) is discouraged by the stiff fees mutual funds charge every investor for buying and then selling shares on the same day.Answer: BTopic: Chapter 20.8 Conflicts of Interest in the Mutual Fund IndustryQuestion Status: Previous Edition31) Late trading and market timingA) allow large, favored investors in a mutual fund to profit at the expense of other investors in the fund.B) hurt ordinary investors by increasing the number of fund shares and diluting the fund's net asset value.C) are both A and B of the above.D) are none of the above.Answer: CTopic: Chapter 20.8 Conflicts of Interest in the Mutual Fund Industry Question Status: Previous Edition32) Which of the following is not a proposal to deal with abuses in the mutual fund industry?A) Strictly enforce the 4:00 PM net asset value rule.B) Make redemption fees mandatory.C) Disclose compensation arrangements for investment advisers.D) Increase the number of dependent directors.Answer: DTopic: Chapter 20.8 Conflicts of Interest in the Mutual Fund IndustryQuestion Status: Previous Edition33) ________ means the investors can convert their investment into cash quickly at a low cost.A) Liquidity intermediationB) Denomination intermediationC) DiversificationD) Managerial expertiseAnswer: ATopic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Previous Edition34) At the start of 2014, one share of Berkshire Hathaway's A-shares was trading at over $150,000. ________ in an mutual fund gives a small investor access to these shares.A) Liquidity intermediationB) Denomination intermediationC) DiversificationD) Managerial expertiseAnswer: BTopic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Previous Edition35) Mutual fund companies frequently offer a number of separate mutual funds called ________.A) indexesB) complexesC) componentsD) actuariesAnswer: BTopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition36) Equity funds can be placed in which class according to the Investment Company Institute?A) Capital appreciation fundsB) World fundsC) Total return fundsD) All of the aboveAnswer: DTopic: Chapter 20.4 Investment Objective Classes Question Status: Previous Edition37) Government bonds are essentially default risk-free, ________ returns.A) and will yield highB) and will yield the highestC) but will have relatively lowD) none of the aboveAnswer: CTopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition38) ________ bonds combine stocks into one fund.A) HybridB) Money marketC) MunicipalD) EquityAnswer: ATopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition39) All ________ are open-end investment funds that invest only in money market securities.A) Stock fundsB) Bond fundsC) Money market mutual fundsD) all of the aboveAnswer: CTopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition20.2 True/False1) The larger the number of shares traded in a stock transaction, the lower the transaction costs per share.Answer: TRUETopic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Previous Edition2) The increase in the number of defined contribution pension funds has slowed the growth of mutual funds.Answer: FALSETopic: Chapter 20.1 The Growth of Mutual FundsQuestion Status: Previous Edition3) Mutual funds accounted for $5.3 trillion, or 27%, of the $19.5 trillion U.S. retirement market at the beginning of 2013.Answer: TRUETopic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Updated from Previous Edition4) Among the investors in mutual funds, only about 25% cite preparing for retirement as one of their main reasons for holding shares.Answer: FALSETopic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Updated from Previous Edition5) Open-end mutual funds are more common than closed-end funds.Answer: TRUETopic: Chapter 20.3 Mutual Fund StructureQuestion Status: Previous Edition6) The net asset value of a mutual fund is the average market price of the stocks, bonds, and other assets the fund owns.Answer: FALSETopic: Chapter 20.3 Mutual Fund StructureQuestion Status: Previous Edition7) A mutual fund's board of directors picks the securities that will be held and makes buy and sell decisions.Answer: FALSETopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition8) Money market mutual funds originated when the brokerage firm Merrill Lynch offered its customers an account from which funds could be taken to purchase securities and into which funds could be deposited when securities were sold. Answer: TRUETopic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition9) A deferred load is a fee charged when shares in a mutual fund are redeemed. Answer: TRUETopic: Chapter 20.5 Fee Structure of Investment FundsQuestion Status: Previous Edition10) Several academic research studies show that investors earn higher returns by investing in mutual funds that charge higher fees.Answer: FALSETopic: Chapter 20.5 Fee Structure of Investment FundsQuestion Status: Previous Edition11) Hedge funds have a minimum investment requirement of between $100,000 and$20 million, with the typical minimum investment being $1 million.Answer: TRUETopic: Chapter 20.7 Hedge FundsQuestion Status: New Question12) SEC research suggests that about three-fourths of mutual funds let privileged shareholders engage in market timing.Answer: TRUETopic: Chapter 20.8 Conflicts of Interest in the Mutual Fund IndustryQuestion Status: Previous Edition13) One factor explaining the rapid growth in mutual funds is that they are financial intermediaries that are not regulated by the federal government.Answer: FALSETopic: Chapter 20.1 The Growth of Mutual FundsQuestion Status: Previous Edition14) Whether a fund is organized as a closed- or an open-end fund, is will have the same basic organizational structure.Answer: TRUETopic: Chapter 20.3 Mutual Fund StructureQuestion Status: Previous Edition15) The primary purpose of loads is to provide compensation for sales brokers. Answer: TRUETopic: Chapter 20.5 Fee Structure of Investment FundsQuestion Status: Previous Edition16) Mutual funds are regulated under four federal laws designed to protect investors. Answer: TRUETopic: Chapter 20.6 Regulation of Mutual FundsQuestion Status: Previous Edition20.3 Essay1) What benefits do mutual funds offer investors?Topic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: Previous Edition2) How is a mutual fund's net asset value calculated?Topic: Chapter 20.3 Mutual Fund StructureQuestion Status: Previous Edition3) How did money market mutual funds originate and why did they become especially popular in the late 1970s and early 1980s?Topic: Chapter 20.1 The Growth of Mutual FundsQuestion Status: Previous Edition4) How does the governance structure of mutual funds lead to asymmetric information and conflicts of interest?Topic: Chapter 20.8 Conflicts of Interest in the Mutual Fund IndustryQuestion Status: Previous Edition5) Describe the practices of late trading and market timing and explain how these practices harm a mutual fund's shareholders.Topic: Chapter 20.8 Conflicts of Interest in the Mutual Fund IndustryQuestion Status: Previous Edition6) Discuss the proposals that have been made to reduce the conflict of interest abuses in the mutual funds industry.Topic: Chapter 20.8 Conflicts of Interest in the Mutual Fund IndustryQuestion Status: Previous Edition7) How is an index fund different from the other four primary investment objective classes for mutual funds?Topic: Chapter 20.4 Investment Objective ClassesQuestion Status: New Question8) Discuss the four primary classes of mutual funds available to investors.Topic: Chapter 20.4 Investment Objective ClassesQuestion Status: Previous Edition9) What are the five benefits of mutual funds?Topic: Chapter 20.2 Benefits of Mutual FundsQuestion Status: New Question10) What is the difference between an open-end and a closed-end mutual fund? Topic: Chapter 20.3 Mutual Fund StructureQuestion Status: New Question11) What are two key differences between a traditional mutual fund and a hedge fund?Topic: Chapter 20.7 Hedge FundsQuestion Status: New Question。
Duration Modeling Techniques for Continuous Speech RecognitionJanne Pylkk¨o nen and Mikko KurimoNeural Networks Research CentreHelsinki University of Technology,Finlandjanne.pylkkonen@hut.fi,mikko.kurimo@hut.fiAbstractPhone durations play a significant part in the com-prehension of speech.The duration information is still mostly disregarded in automatic speech recognizers due to the use of hidden Markov models(HMMs)which are deficient in modeling phone durations properly.Previous results have shown that using different approaches for ex-plicit duration modeling have improved the isolated word recognition in English.However,a unified comparison between the methods has not been reported.In this paper three techniques for explicit duration modeling are compared and evaluated in a large vocab-ulary continuous speech recognition task.The target lan-guage was Finnish,in which phone durations are espe-cially important for proper understanding.The results show that the choice of the duration modeling technique depends on the speed requirements of the recognizer.The best technique required a slightly longer running time than without an explicit duration model,but achieved an 8%relative improvement to the letter error rate.1.IntroductionThe modern automatic speech recognition(ASR)sys-tems are based on modeling the phones with hidden Markov models(HMM),using HMM states in a left-to-right topology for each phone.The transition probabili-ties of the HMMs represent the statistical duration infor-mation of the phones.It has been noted that these transi-tion probabilities have little effect to the recognition per-formance[1],and hence it is customary to ignore the use of more detailed durational information and rely more on the actual acoustic data.The durational information is still worth of further examination.Although phone durations do not have ac-tual discriminative role in English,they do help in distin-guishing several words from each other,such as sit and seat or ship and sheep.In some other languages,for ex-ample in Finnish,phone durations can be the only clue in discriminating between certain words.Good duration modeling can therefore be a major issue.It has been reported in several papers that using ex-plicit state duration models with hidden Markov mod-els improve the recognition accuracy[2,3,4].However,most of the evaluations in these papers have been isolated word recognition tests with connected word models,not continuous speech recognition tests with phoneme based models nowadays in use.Besides,no single method have been found which would completely satisfy the modeling needs,and the different approaches have varying implica-tions,for example,to the recognition efficiency.To gain more insight into this matter,this paper presents a com-parison between three different extensions to integrate ex-plicit duration models into the HMMs.The modeling techniques are evaluated using a modern phoneme based ASR[5]in a large vocabulary continuous speech recog-nition(LVCSR)task.2.HMM based duration modelingtechniquesIncorporating explicit state duration models into the HMMs introduces problems,as it breaks up some of the assumptions which are employed in the efficient HMM algorithms.A direct consequence of the Markov assump-tion is that state durations have a geometric distribution, defined by the probability of the self-transition.When this distribution is replaced with an explicitly defined one,the Markov assumption no longer holds.The Baum-Welch and Viterbi algorithms[6]used tofind the opti-mal paths through an HMM heavily depend on this as-sumption,so they are no longer applicable in their basic forms.Modifying them to properly deal with the loss of this simplifying assumption seriously degrades their effi-ciency.The solution is then tofind some other restrictive assumptions or to use sub-optimal algorithms.Before reviewing the different duration modeling techniques,the distributions of phone durations arefirst examined.2.1.Phone durations distribution modelsFor a phone model with three HMM states,the prior distribution of a phone duration is the convolution of three geometric distributions determined by the transition probabilities of the HMM.The properties of this prior distribution can be analyzed by considering the state du-rations as independent random variables.The mean andFigure2:Sub-HMM topology.The rightmost state illus-trates the next HMM or sub-HMM state,so that the mini-mum duration in the sub-HMM is one.imum duration of one normal HMM state.Expanding each HMM state to this kind of sub-HMM introduces a large number of free parameters to be es-timated.It may be therefore necessary to constrain the parameters in some way.In[10]the number of states in sub-HMMs were determined by the number of occur-rences in the HMM state in the training phase,and the transition probabilities of all the sub-HMM states of all the phone models were set to be the same.In this work, a heuristic rule for determining the number of sub-HMM states was used so that goodfits to the measured dura-tion distributions were achieved with low numbers of sub-HMM states.On average,the sub-HMMs had3.8states. The transition probabilities were constrained byfitting a gamma distribution to the measured duration distribution.2.4.Post-processor duration modelBoth HSMM and ESHMM degrade the efficiency of the recognition,the former by altering the algorithms and the latter by introducing additional states for the HMMs. Juang et al.[2]proposed a duration model which avoids this kind of loss of efficiency.Their method uses the out-put of the Viterbi algorithm and ranks the proposed paths using better models for the state durations.The method is therefore called the post-processor duration model.The augmentation of the log likelihood given by the Viterbi algorithm can be stated aslogˆf=log f+αNj=1log d j(τj).(1)f denotes the likelihood score given by the Viterbi search,αis an empirical scaling factor,N is the number of dis-tinct HMM states through which the best path traversed, d j are the duration probability distribution functions of those states,andτj are the durations spent in each state.3.Evaluation3.1.SetupThe utility of the duration modeling techniques was eval-uated with speaker dependent speech recognition tests. Finnish was used as the target language,as the proper un-derstanding of Finnish is more dependent on the correct durational information than,for example,with English. The speech material was a book spoken by a professional speaker,which was a reasonable choice for minimizing the unwanted variation of phone durations.An extract of 12hours was used to train the models,and independent parts of9and30minutes were used as development and evaluation sets,respectively.The development set was used to optimize the empirical scaling factors for the log likelihoods of the language model,the transition proba-bilities,and the duration distribution probabilities.The speech recognition system used for the evaluation has been presented in[5].The number of triphone mod-els was empirically adjusted to the available data.For the language model,a morph based trigram model was used. All the duration distributions were modeled with gamma distributions.As the different duration modeling tech-niques affect both the efficiency and the accuracy of the recognition,the recognition tests were run with different pruning settings(affecting the optimality of the Viterbi algorithm)to achieve different running times.The run-ning time is indicated by a real-time factor,which should be interpreted only as a relative value for the number of reasons affecting the actual speed of the recognition.The recognition accuracy was measured by a letter error rate (LER).As compared with the word error rate(WER),it is more suitable for a language such as Finnish where rather long words consisting of many morphemes are common.3.2.ResultsFigure3shows the recognition accuracy as a function of the real-time factor for different setups.The model la-beled as“HMM”is the baseline result without explicit duration modeling.Thefigure shows clearly that the in-tended running speed affects the choice of the best du-ration modeling technique.For moderate speeds(real-time factors10to30)the post-processor model functions best.But if the pruning level of the recognition is set to low enough,the HSMM outperforms the others.The ESHMM does not seem to produce good results,despite its intuitive approach.All the models suffer from randomfluctuation in the LER measurements after they have reached their optimal running speed.This is due to inherent noise in the mea-surements,along with the effects resulting from the gen-eral pruning strategy used in the Viterbi algorithm.Mea-sured from the points of the best performance,the let-ter error rate of the post-processor duration model was 2.73%,corresponding to the word error rate of15.3%. The HSMM achieved a LER of2.63%(WER15.2%). Compared to the baseline result with a LER of2.88% (WER16.2%),the post-processor duration model im-proved the LER about5%,while the HSMM achieved about8%relative improvement.。
ARK方舟生存进化代码大全方舟指令按~ 进入控制台**24号补丁以后进入控制台改按TAB黑色是命令,绿色是中英说明-------------------------------------------------------------------------------多人模式---------------------------------------------------------SetCheatPlayer True Enables Cheat Menu 开启作弊选单SetCheatPlayer False Disables Cheat Menu 关闭作弊选单AdminCheat AllowPlayerToJoinNoCheck Whitelists a player 将玩家设为白名单AdminCheat DisallowPlayerToJoinNoCheck Removes player from whitelist 将玩家从白名单移除AdminCheat Broadcast Broadcasts a message to the whole server. 送出信息给所有人AdminCheat SetMessageOfTheDay Sets the MOD 设定MOD AdminCheat God Unkillable, except you can drown. 无敌AdminCheat Fly Able to Fly 飞行模式AdminCheat Walk Deactivates Flying 取消飞行模式AdminCheat Teleport Teleports you in the direction you are facing. 传送至你当前看的点AdminCheat Slomo <#> Changes server speed. Values are 1-5. 1 = Normal Speed 设定时间速度<1-5> ,1=正常速度AdminCheat PlayersOnly Freezes all Dinos at their current position. Also freezes crafting. 冻结所有恐龙行动,制作中的工作也会受影响AdminCheat Ghost Noclip, walk through walls/objects. 穿墙模式AdminCheat ForceTame Instantly tames a Dino. Can ride without saddle. 强制驯化恐龙(看著要驯的恐龙打这指令)AdminCheat AddExperience 1000 0 0 1 Gives yourself 1000 XP 给自己经验值1000(可改其它值)-AddExperience (value for first number) (toggle option in relation to tribe) (Second toggle) (third toggle) so in total there should be 4 separate numbers. AdminCheat GiveResources Gives you 50 of all resources 给各种资源物件50份AdminCheat InfiniteStats Infinite Hunger, Stamina, Ammo, etc.. 状态/子弹不减(饥饿、耐力、免装弹....等)BanPlayer Bans user from server 后面加玩家ID,将玩家踢出UnBanPlayer Unbans user from server 后面加玩家ID, 将玩家取消踢出AdminCheat DamageTarget Damages a creature you are looking at for a set amount 伤害当前所看的目标AdminCheat DestroyAllEnemies Destroys all enemies. They respawn after a while. 杀死所有的恐龙,过一会会重生出来GiveEngrams unlocks all crafting recipes for your character -* Bugged - You can't craft these recipes 开启所有蓝图(目前此功能有问题) AdminCheat HurtMe Deals damage to yourself 伤害自己(自杀)ToggleGun Toggles visibility of current equipped item 当前装备物品隐形开/关AdminCheat SetTimeOfDay Changes time of day 设定时间(后面要加时间如settimeofday 04:00)AdminCheat SetPlayerPos 0 0 0 Allows you to teleport to coordinates. 传送到座标( 如setplayerpos 12,12,132)AdminCheat SaveWorld Saves current worldstate 手动储存AdminCheat Quit Exits the current world. Use after saving for a safe shutdown. 离开游戏AdminCheat ExecSetSleeping True/False Puts character to sleep/wakes them up 设定玩家为睡眠(true)/醒来(false)AdminCheat EnemyInvisible True/False Makes all creatures ignoreyou even if you attack them 玩家隐身,恐龙就算被打也看不到AdminCheat DestroyAll Destroys all Objects/Dinos of a classname 删除所有物件及恐龙AdminCheat Summon Summons a dino at your location. 召唤一只恐龙到当前位置AdminCheat GiveItemNum Gives you an item. 给指定的物品请参照下面的物品ID >>例1:giveitemnum 105 1 1 false 给储存箱1个Stat FPS Shows your FPS and latency, usable by anyone. 显示帧数--------------------------------------------------------------------------------------------------------单人模式-----------------------------------------------------SetCheatPlayer True Enables Cheat Menu 开启作弊选单SetCheatPlayer False Disables Cheat Menu 关闭作弊选单God Unkillable, except you can drown. 无敌Fly Able to Fly 飞行模式Walk Deactivates Flying 取消飞行模式Teleport Teleports you in the direction you are facing. 传送至你当前看的点Slomo <#> Changes server speed. Values are 1-5. 1 =Normal Speed 设定时间速度<1-5> ,1=正常速度PlayersOnly Freezes all Dinos at their current position. Also freezes crafting. 冻结所有恐龙行动,制作中的工作也会受影响Ghost Noclip, walk through walls/objects. 穿墙模式ForceTame Instantly tames a Dino. Can ride without saddle. 强制驯化恐龙(看著要驯的恐龙打这指令)AddExperience 1000 0 0 1 Gives yourself 1000 XP 给自己经验值1000(可改其它值)-AddExperience (value for first number) (toggle option in relation to tribe) (Second toggle) (third toggle) so in total there should be 4 separate numbers. GiveResources Gives you 50 of all resources 给各种资源物件50份InfiniteStats Infinite Hunger, Stamina, Ammo, etc.. 状态/子弹不减(饥饿、耐力、免装弹....等)DamageTarget Damages a creature you are looking at for a set amount 伤害当前所看的目标DestroyAllEnemies Destroys all enemies. They respawn after a while. 杀死所有的恐龙,过一会会重生出来GiveEngrams unlocks all crafting recipes for your character -* Bugged - You can't craft these recipes 开启所有蓝图(目前此功能有问题) HurtMe Deals damage to yourself 伤害自己(自杀)例hurtme 1000ToggleGun Toggles visibility of current equipped item 当前装备物品隐形开/关SetTimeOfDay Changes time of day 设定时间(后面要加时间如settimeofday 04:00)SetPlayerPos 0 0 0 Allows you to teleport to coordinates. 传送到座标( 如setplayerpos 12,12,132)SaveWorld Saves current worldstate 手动储存Quit Exits the current world. Use after saving for a safe shutdown. 离开游戏ExecSetSleeping True/False Puts character to sleep/wakes them up 设定玩家为睡眠(true)/醒来(false)EnemyInvisible True/False Makes all creatures ignore you even if you attack them 玩家隐身,恐龙就算被打也看不到DestroyAll Destroys all Objects/Dinos of a classname 删除所有物件及恐龙Summon Summons a dino at your location. 召唤一只恐龙到当前位置GiveItemNum Gives you an item. 给指定的物品请参照下面的物品ID >>例1:giveitemnum 105 1 1 false 给储存箱1个Stat FPS Shows your FPS and latency, usable by anyone. 显示帧数cheat engine(CE) 修改表(27号补丁测试有效) ARK.CT (50.08 KB, 下载次数: 953)Full Player Status 玩家状态全满Player Pointer 各项属性(手动修改)Unlimited Ammo (untested) 无限子弹Unlimited Inventory 物品不减Unlimited Durability 耐用度不减Ignore Engram Requirements 忽略技能点需求Ignore Crafting Requirements 忽略蓝图需求Instant Taming 秒驯Unlimited Engram 无限技能点Unlimited Buff Durations 无限状态加成时间Inventory Selection Pointer 选定物品修改全物品IDItem ID Item Description1 Simple Pistol2 Assault Rifle3 Rocket Launcher4 Simple Bullet5 Bow6 Genade7 Wood8 Stone9 Metal10 Hide11 Chitin12 Raw Meat13 Spoiled Meat14 Cooked Meat15 Water Jar16 Water Jar (Full)17 Cloth Pants18 Cloth Shirt19 Cloth Hat20 Cloth Boots21 Cloth Gloves22 Hide pants23 Hide Shirt24 Hide Hat25 Hide Boots26 Hide Gloves27 Chitin Leggings28 Chitin Chestpiece29 Chitin Helmet30 Chitin Boots31 Chitin Gauntlets32 Stone Arrow33 Stone Pick34 Stone Hatchet35 Metal Pick36 Metal Hatchet37 Torch38 Paintbrush39 Campfire40 Standing Torch41 Hide Sleeping Bag42 Remote Detonator43 C4 Charge44 Blood Extraction Syringe45 Blood Pack46 Improvised Explosive Device47 Waterskin48 Waterskin(Full)49 Berrybush Seeds50 Fertilizer51 Bingleberry Soup52 Medical Brew53 Energy Brew54 Dinosaur Feces55 Human Feces56 Stegosaurus Egg57Spear58 Red Coloring59 Green Coloring60 Blue Coloring61 Yellow Coloring62 Purple Coloring63 Orange Coloring64 Black Coloring65 White Coloring66 Brown Coloring67 Cyan Coloring68 Purple Coloring69 Rex Saddle70 Tranq Arrow71 Pistol Hat Skin72 GPS73 Flint74 Metal Ingot75 Thatch76 Fiber77 Charcoal78 Crystal79 Thatch Roof80 Thatch Door81 Thatch Foundation82 Thatch Wall83 Thatch Doorframe84 Wooden Catwalk85 Wooden Ceiling86 Wooden Hatchframe87 Wooden Door88 Wooden Foundation89 Wooden Ladder90 Wooden Pillar91 Wooden Ramp92 Wooden Trapdoor93 Wooden Wall94 Wooden Doorframe95 Wooden WindowFrame96 Wooden Window97 Wooden Sign98 Blueprint: Note99 Citronal100 Parasaur Saddle101 Raptor Saddle102 Stego Saddle103 Trike Saddle104 Pulmonoscorpius Saddle105 Storage Box106 Large Storage Box107 Mortar and Pestle108 Sparkpowder109 Gunpowder110 Stone Irrigation Pipe - Intake111 Stone Irrigation Pipe - Straight 112 Stone Irrigation Pipe - Incline113 Stone Irrigation Pipe - Intersection 114 Stone Irrigation Pipe - Vertical 115 Stone Irrigation Pipe - Tap116 Amarberry Seed117 Amarberry118 Azulberry119 Tintoberry120 Mejoberry121 Narcoberry122 Stimberry123 Narcotic124 Stimulant125 Refining Forge126 Smithy127 Compost Bin128 Cooking Pot129 Simple Bed130 Small Crop Plot131 Pteranodon Saddle132 Longneck Rifle133 Citronal Seed134 Specimen Implant135 Bronto Saddle136 Wooden Fence Foundation 137 Compass138 Scope Attachment139 Slingshot140 Pike141 Radio142 Obsidian143 Dinosaur Gateway144 Simple Rifle Ammo145 Summon Broodmother 146 Cementing Paste147 Dinosaur Gate148 Artifact of the Hunter149 Artifact of the Pack150 Artifact of the Massive 151 Artifact of the Devious 152 Artifact of the Clever153 Artifact of the Skylord154 Artifact of the Devourer 155 Artifact of the Queen156 Artifact of the Strong157 Artifact of the Flamekeeper 158 Argentavis Talon159 Megalodon Tooth160 Tyrannosaur Arm161 Saurpod Vertebra162 Oil163 Silica Pearls164 Gasoline165 Electronics166 Polymer167 Metal Catwalk168 Metal Ceiling169Metal Hatchframe 170 Metal Door171 Metal Fence Foundation172 Metal Foundation173 Behemoth Gate174 Behemoth Gateway175 Metal Ladder176 Metal Pillar177 Metal Ramp178 Metal Trapdoor179 Metal Wall180 Metal Doorframe181 Metal Windowframe182 metal Window183 Super Test meat184 Flare Gun185 Fabricator186 Water Tank187 Parachute188 Air Conditioner189 Electrical Generator190 Electrical Outlet191 Inclined Electrical Cable192 Electrical Cable Intersection193 Straight Electrical Cable194 Vertical Electrical Cable195 Lamppost196 Refrigerator197 Auto Turret198 Remote Keypad199 Metal Irrigation Pipe - Inclined 200 Metal Irrigation Pipe - Tap201 Metal Irrigation Pipe - Intersection 202 Metal irrigation Pipe - Straight 203 Metal Irrigation Pipe - Tap204 Metal Irrigation Pipe - Vertical 205 Flashlight Attachment206 Silencer Attachment 207 Carbonemys Saddle 208 Sarco Saddle209 Ankylo Saddle210 Mammoth Saddle211 Megalodon Saddle 212 Sabertooth Saddle 213 Carno Saddle214 Argentavis Saddle215 Plesiosaur Saddle216 Chitin/Keratin217 Keratin218 Metal Sign219 Holo-Scope Attachment 220 Laser Attachment221 Wooden Billboard222 Flak Leggings223 Flak Chestpiece224 Flak Helmet225 Flak Boots226 Flak Gauntlets227 Enduro Stew228 Lazarus Chowder229 Calien Soup230 Fria Curry231 Focal Chilli232 Savoroot233 Longgrass1234 Rockarrot235 Azulberry Seed236 Tintoberry Seed237 Mejoberry Seed238 Narcoberry Seed239 Stimberry Seed240 Savoroot Seed241 Longgrass Seed242 Rockarrot Seed243 Metal Billboard244 Fabricated Pistol245 Advanced Bullet246 Advanced Rifle Bullet 247 Rocket Propelled Grenade 248 Medium Crop Plot249 Large Crop Plot250 Rare Flower251 Rare Mushroom252 Raw Prime Meat253 Cooked Prime Meat254 Battle Tartare255 Shadow Steak Saute256 Rockwell Recipes: Enduro Stew257 Rockwell Recipes: Lazarus Chowder 258 Rockwell Recipes: Calien Soup259 Rockwell Recipes: Fria Curry260 Rockwell Recipes: Focal Chilli261 Rockwell Recipes: Battle Tartare262 Rockwell Recipes: Shadow Steak Saute 263 Notes on Rockwell Recipes264 Wall Sign265 Metal Dinosaur Gateway266 Metal Dinosaur Gate267 Shotgun268 Simple Shotgun Ammo269 Metal WallSign270 Amarberry Seed271 Azulberry Seed272 Tintoberry Seed273 Narcoberry Seed274 Stimberry Seed275 Mejoberry Seed276 Citronal Seed277 Savaroot Seed278 Longgrass Seed279 Rockarrot Seed280 Flag281 Hunter Skin Hat282 Rex Stomped Glasses Saddle 283 Spyglass284 Spider Flag285 Phiomia Saddle286 Medium Dinosaur Feces 287 Large Dinosaur Feces288 Brontosaurus Egg289 Parasaur Egg290 Raptor Egg291 T-Rex Egg292 Triceratops Egg293 Parasaur ARK Founder Saddle Skin294 Dodo Egg---------------------------------------------------------------------------------------------------------------------------------------------------------------------全召唤陆上型Ankylo summon Ankylo_Character_BP_CAraneo summon SpiderS_Character_BP_CCarno summon Carno_Character_BP_CDilo summon Dilo_Character_BP_CDodo summon Dodo_Character_BP_CMammoth summon Mammoth_Character_BP_CParasaur summon Para_Character_BP_CPhioma summon Phiomia_Character_BP_CRaptor summon Raptor_Character_BP_CSaber summon Saber_Character_BP_CSarco summon Sarco_Character_BP_CBrontosaurus summon Sauropod_Character_BP_C Scorpion summon Scorpion_Character_BP_C Stegosaurus summon Stego_Character_BP_C Titanoboa summon BoaFrill_Character_BP_CT-Rex summon Rex_Character_BP_CTrike summon Trike_Character_BP_CTurtle summon Turtle_Character_BP_C水中型Coel(Fish) summon Coel_Character_BP_C Megalodon summon Megalodon_Character_BP_C Piranha summon Piranha_Character_BP_C空中型Argentavis summon Argent_Character_BP_COnyc summon Bat_Character_BP_CPtero summon Ptero_Character_BP_CBOSS型Broodmother Lyrix summon SpiderL_Character_BP_C。
US president George W. Bush knows he’s unpopular. But he thinks he has done his best to solve the financial crisis, in which "… the excesses of the past have caused a lot of folks to hurt when it comes to their 401(k)’s or jobs," Bush said in an interview with Fox News Channel last week. What is a 401(k)? This is the name of a retirement savings plan initiated by the US federal government. Employees can choose to put a certain amount of their salary into an account and withdraw the money when they retire. The money is invested collectively. The financial crisis has shrunk many people’s 401k funds. The fund’s name comes from the section of US law that initiated it: the 401(k) section of the US Internal Revenue Code, passed into law in 1978. Here are some other numbers that carry special significance in American English: 101 Pronounced "one-oh-one", it refers to courses intended for beginners – "English 101" (英语⼊门课), for example. The university offers 101 courses for those freshmen and sophomores who have little knowledge of mathematics. 411 When someone says 411, they are referring to information about something; 411 is the number Americans dial to find people’s telephone numbers. The number 114 is used for this service in China. So what’s the 411 on your recent job interview? [What’s the news about your recent job interview?] 9 times out of 10 = most of the time Nine times out of 10, she will cook a delicious meal. put 2 and 2 together = think about something I think if you put two and two together you will make the right decision.。
Financial Markets and Institutions, 8e (Mishkin)Chapter 14 The Mortgage Markets14.1 Multiple Choice1) Which of the following are important ways in which mortgage markets differ from the stock and bond markets?A) The usual borrowers in the capital markets are government entities and businesses, whereas the usual borrowers in the mortgage markets are individuals.B) Most mortgages are secured by real estate, whereas the majority of capital market borrowing is unsecured.C) Because mortgages are made for different amounts and different maturities, developing a secondary market has been more difficult.D) All of the above are important differences.E) Only A and B of the above are important differences.Answer: DTopic: Chapter 14.1 What Are Mortgages?Question Status: Previous Edition2) Which of the following are important ways in which mortgage markets differ from stock and bond markets?A) The usual borrowers in capital markets are government entities, whereas the usual borrowers in mortgage markets are small businesses.B) The usual borrowers in capital markets are government entities and large businesses, whereas the usual borrowers in mortgage markets are small businesses.C) The usual borrowers in capital markets are government entities and large businesses, whereas the usual borrowers in mortgage markets are small businesses and individuals.D) The usual borrowers in capital markets are businesses and government entities, whereas the usual borrowers in mortgage markets are individuals.Answer: DTopic: Chapter 14.1 What Are Mortgages?Question Status: Previous Edition3) Which of the following are true of mortgages?A) A mortgage is a long-term loan secured by real estate.B) A borrower pays off a mortgage in a combination of principal and interest payments that result in full payment of the debt by maturity.C) Over 80 percent of mortgage loans finance residential home purchases.D) All of the above are true of mortgages.E) Only A and B of the above are true of mortgages.Answer: DTopic: Chapter 14.1 What Are Mortgages?Question Status: Previous Edition4) Which of the following are true of mortgages?A) A mortgage is a long-term loan secured by real estate.B) Borrowers pay off mortgages over time in some combination of principal and interest payments that result in full payment of the debt by maturity.C) Less than 65 percent of mortgage loans finance residential home purchases.D) All of the above are true of mortgages.E) Only A and B of the above are true of mortgages.Answer: ETopic: Chapter 14.1 What Are Mortgages?Question Status: Previous Edition5) Which of the following are true of mortgage interest rates?A) Interest rates on mortgage loans are determined by three factors: current long-term market rates, the term of the mortgage, and the number of discount points paid.B) Mortgage interest rates tend to track along with Treasury bond rates.C) The interest rate on 15-year mortgages is lower than the rate on 30-year mortgages, all else the same.D) All of the above are true.E) Only A and B of the above are true.Answer: DTopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition6) Which of the following are true of mortgages?A) More than 80 percent of mortgage loans finance residential home purchases.B) The National Banking Act of 1863 rewarded banks that increased mortgage lending.C) Most mortgages during the 1920s and 1930s were balloon loans.D) All of the above are true.E) Only A and C of the above are true.Answer: ETopic: Chapter 14.1 What Are Mortgages?Question Status: Previous Edition7) Which of the following is true of mortgage interest rates?A) Longer-term mortgages have lower interest rates than shorter-term mortgages.B) Mortgage rates are lower than Treasury bond rates because of the tax deductibility of mortgage interest rates.C) In exchange for points, lenders reduce interest rates on mortgage loans.D) All of the above are true.E) Only A and B of the above are true.Answer: CTopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition8) Typically, discount points should not be paid if the borrower will pay off the loan in ________ years or less.A) 5B) 10C) 15D) 20Answer: ATopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition9) Which of the following is true of mortgage interest rates?A) Longer-term mortgages have higher interest rates than shorter-term mortgages.B) In exchange for points, lenders reduce interest rates on mortgage loans.C) Mortgage rates are lower than Treasury bond rates because of the tax deductibility of mortgage interest payments.D) All of the above are true.E) Only A and B of the above are true.Answer: ETopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition10) Which of the following reduces moral hazard for the mortgage borrower?A) CollateralB) Down paymentsC) Private mortgage insuranceD) Borrower qualificationsAnswer: BTopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition11) Which of the following protects the mortgage lender's right to sell property if the underlying loan defaults?A) A lienB) A down paymentC) Private mortgage insuranceD) Borrower qualificationE) AmortizationAnswer: ATopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition12) Which of the following is true of mortgage interest rates?A) Mortgage rates are closely tied to Treasury bond rates, but mortgage rates tend to stay below Treasury rates because mortgages are secured with collateral.B) Longer-term mortgages have higher interest rates than shorter-term mortgages.C) Interest rates are higher on mortgage loans on which lenders charge points.D) All of the above are true.E) Only A and B of the above are true.Answer: BTopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition13) During the early years of an amortizing mortgage loan, the lender appliesA) most of the monthly payment to the outstanding principal balance.B) all of the monthly payment to the outstanding principal balance.C) most of the monthly payment to interest on the loan.D) all of the monthly payment to interest on the loan.E) the monthly payment equally to interest on the loan and the outstanding principal balance.Answer: CTopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition14) During the last years of an amortizing mortgage loan, the lender appliesA) most of the monthly payment to the outstanding principal balance.B) all of the monthly payment to the outstanding principal balance.C) most of the monthly payment to interest on the loan.D) all of the monthly payment to interest on the loan.E) the monthly payment equally to interest on the loan and the outstanding principal balance.Answer: ATopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition15) During the last years of a balloon mortgage loan, the lender appliesA) most of the monthly payment to the outstanding principal balance.B) all of the monthly payment to the outstanding principal balance.C) most of the monthly payment to interest on the loan.D) all of the monthly payment to interest on the loan.E) the monthly payment equally to interest on the loan and the outstanding principal balance.Answer: DTopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition16) During the early years of a balloon mortgage loan, the lender appliesA) most of the monthly payment to the outstanding principal balance.B) all of the monthly payment to the outstanding principal balance.C) most of the monthly payment to interest on the loan.D) all of the monthly payment to interest on the loan.E) the monthly payment equally to interest on the loan and the outstanding principal balance.Answer: DTopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition17) A borrower who qualifies for an FHA or VA loan enjoys the advantage thatA) the mortgage payment is much lower.B) only a very low or zero down payment is required.C) the cost of private mortgage insurance is lower.D) the government holds the lien on the property.Answer: BTopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition18) (I) Conventional mortgages are originated by private lending institutions, and FHA or VA loans are originated by the government. (II) Conventional mortgages are insured by private companies, and FHA or VA loans are insured by the government.A) (I) is true, (II) false.B) (I) is false, (II) true.C) Both are true.D) Both are false.Answer: BTopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition19) Borrowers tend to prefer ________ to ________, whereas lenders prefer ________.A) fixed-rate loans; ARMs; fixed-rate loansB) ARMs; fixed-rate loans; fixed-rate loansC) fixed-rate loans; ARMs; ARMsD) ARMs; fixed-rate loans; ARMsAnswer: CTopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition20) (I) ARMs offer lower initial rates and the rate may fall during the life of the loan. (II) Conventional mortgages do not allow a borrower to take advantage of falling interest rates.A) (I) is true, (II) is false.B) (I) is false, (II) is true.C) Both are true.D) Both are false.Answer: ATopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition21) Growing-equity mortgages (GEMs)A) help the borrower pay off the loan in a shorter time.B) have such low payments in the first few years that the principal balance increases.C) offer borrowers payments that are initially lower than the payments on aconventional mortgage.D) do all of the above.E) do only A and B of the above.Answer: ATopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition22) A borrower with a 30-year loan can create a GEM byA) simply increasing the monthly payments beyond what is required and designating that the excess be applied entirely to the principal.B) converting his ARM into a conventional mortgage.C) converting his conventional mortgage into an ARM.D) converting his conventional mortgage into a GPM.Answer: ATopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition23) Which of the following are useful for home buyers who expect their income to rise in the future?A) GPMsB) RAMsC) GEMsD) Only A and B are useful.E) Only A and C are useful.Answer: ETopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition24) Which of the following are useful for home buyers who expect their income to fall in the future?A) GPMsB) RAMsC) GEMsD) Only A and B are useful.E) Only A and C are useful.Answer: BTopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition25) Retired people can live on the equity they have in their homes by using aA) GEM.B) GPM.C) SAM.D) RAM.Answer: DTopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition26) Second mortgages serve the following purposes:A) they give borrowers a way to use the equity they have in their homes as security for another loan.B) they allow borrowers to get a tax deduction on loans secured by their primary residence or vacation home.C) they allow borrowers to convert their conventional mortgages into GEMs.D) all of the above.E) only A and B of the above.Answer: ETopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition27) Which of the following is a disadvantage of a second mortgage compared to credit card debt?A) The loans are secured by the borrower's home.B) The borrower gives up the tax deduction on the primary mortgage.C) The borrower must pay points to get a second mortgage loan.D) The borrower will find it more difficult to qualify for a second mortgage loan.Answer: ATopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition28) The share of the mortgage market held by savings and loans isA) over 50 percent.B) approximately 40 percent.C) approximately 20 percent.D) less than 5 percent.Answer: DTopic: Chapter 14.4 Mortgage-Lending InstitutionsQuestion Status: Updated from Previous Edition29) The share of the mortgage market held by commercial banks is approximatelyA) 50 percent.B) 30 percent.C) 15 percent.D) 5 percent.Answer: BTopic: Chapter 14.4 Mortgage-Lending Institutions Question Status: Updated from Previous Edition30) A loan-servicing agent willA) package the loan for an investor.B) hold the loan in their investment portfolio.C) collect payments from the borrower.D) do both A and C of the above.E) do both B and C of the above.Answer: CTopic: Chapter 14.5 Loan ServicingQuestion Status: Previous Edition31) Distinct elements of a mortgage loan includeA) origination.B) investment.C) servicing.D) all of the above.E) only B and C of the above.Answer: DTopic: Chapter 14.6 Secondary Mortgage MarketQuestion Status: Previous Edition32) The Federal National Mortgage Association (Fannie Mae)A) was set up to buy mortgages from thrifts so that these institutions could make more loans.B) funds purchases of mortgages by selling bonds to the public.C) provides insurance for certain mortgage contracts.D) does all of the above.E) does only A and B of the above.Answer: ETopic: Chapter 14.6 Secondary Mortgage MarketQuestion Status: Previous Edition33) The Federal Housing Administration (FHA)A) was set up to buy mortgages from thrifts so that these institutions could make more loans.B) funds purchases of mortgages by selling bonds to the public.C) provides insurance for certain mortgage contracts.D) does all of the above.E) does only A and B of the above.Answer: CTopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition34) ________ issues participation certificates, and ________ provides federal insurance for participation certificates.A) Freddie Mac; Freddie MacB) Freddie Mac; Ginnie MaeC) Ginnie Mae; Freddie MacD) Ginnie Mae; Ginnie MaeE) Freddie Mac; no oneAnswer: ETopic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: Previous Edition35) REMICs are most likeA) Freddie Mac pass-through securities.B) Ginnie Mae pass-through securities.C) participation certificates.D) collateralized mortgage obligations.Answer: DTopic: Chapter 14.8 What Is a Mortgage-Backed Security? Question Status: Previous Edition36) Ginnie MaeA) insures qualifying mortgages.B) insures pass-through certificates.C) insures collateralized mortgage obligations.D) does only A and B. of the above.E) does only B and C of the above.Answer: BTopic: Chapter 14.8 What Is a Mortgage-Backed Security? Question Status: Previous Edition37) Mortgage-backed securitiesA) have been growing in popularity in recent years as institutional investors look for attractive investment opportunities.B) are securities collateralized by a pool of mortgages.C) are securities collateralized by both insured and uninsured mortgages.D) are all of the above.E) are only A and B of the above.Answer: DTopic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: Previous Edition38) The most common type of mortgage-backed security isA) the mortgage pass-through, a security that has the borrower's mortgage payments pass through the trustee before being disbursed to the investors.B) collateralized mortgage obligations, a security which reduces prepayment risk.C) the participation certificate, a security which passes the borrower's mortgage payments equally among all the owners of the certificates.D) the securitized mortgage, a security which increases the liquidity of otherwise illiquid mortgages.Answer: ATopic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: Previous Edition39) The interest rate borrowers pay on their mortgages is determined byA) current long-term market rates.B) the term.C) the number of discount points.D) all of the above.Answer: DTopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition40) A loan for borrowers who do not qualify for loans at the usual market rate of interest because of a poor credit rating or because the loan is larger than justified by their income isA) a subprime mortgage.B) a securitized mortgage.C) an insured mortgage.D) a graduated-payment mortgage.Answer: ATopic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: Previous Edition41) The percentage of the total loan paid back immediately when a mortgage loan is obtained, which lowers the annual interest rate on the debt, is calledA) discount points.B) loan terms.C) collateral.D) down payment.Answer: ATopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition42) Which of the following terms are found in mortgage loan contracts to protect the lender from financial loss?A) CollateralB) Down paymentC) Private mortgage insuranceD) All of the aboveAnswer: DTopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition43) What factors are used in determining a person's FICO score?A) Past payment historyB) Outstanding debtC) Length of credit historyD) All of the aboveAnswer: DTopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition44) Between 2000 and 2005, home prices increased an average of ________ per year.A) 2%B) 4%C) 8%D) 12%Answer: CTopic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: New Question45) From 2000 to 2005, housing prices increased, on average, by over 40%. This run up in prices was caused byA) speculators.B) an increase in subprime loans, which increased demand for new and existing houses.C) both A and B.D) None of the above are correct.Answer: CTopic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: Updated from Previous Edition14.2 True/False1) In 2012, mortgage loans to farms represented the largest proportion of mortgage lending in the U.S.Answer: FALSETopic: Chapter 14.1 What Are Mortgages?Question Status: New Question2) Down payments are designed to reduce the likelihood of default on mortgage loans.Answer: TRUETopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition3) Discount points (or simply points) are interest payments made at the beginning of a loan.Answer: TRUETopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition4) A point on a mortgage loan refers to one monthly payment of principal and interest.Answer: FALSETopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition5) Closing for a mortgage loan refers to the moment the loan is paid off.Answer: FALSETopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition6) Private mortgage insurance is a policy that guarantees to make up any discrepancy between the value of the property and the loan amount, should a default occur.Answer: TRUETopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition7) During the early years of a mortgage loan, the lender applies most of the payment to the principal on the loan.Answer: FALSETopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition8) One important advantage to a borrower who qualifies for an FHA or VA loan is the very low interest rate on the mortgage.Answer: FALSETopic: Chapter 14.3 Types of Mortgages9) Adjustable-rate mortgages generally have lower initial interest rates than fixed-rate mortgages.Answer: TRUETopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition10) Mortgage interest rates loosely track interest rates on three-month Treasury bills.Answer: FALSETopic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition11) An advantage of a graduated-payment mortgage is that borrowers will qualify for a larger loan than if they requested a conventional mortgage.Answer: TRUETopic: Chapter 14.3 Types of Mortgages12) Nearly half the funds for mortgage lending comes from mortgage pools and trusts.Answer: FALSETopic: Chapter 14.4 Mortgage-Lending InstitutionsQuestion Status: Updated from Previous Edition13) Many institutions that make mortgage loans do not want to hold large portfolios of long-term securities, because it would subject them to unacceptably high interest-rate risk.Answer: TRUETopic: Chapter 14.4 Mortgage-Lending InstitutionsQuestion Status: Previous Edition14) A problem that initially hindered the marketability of mortgages in a secondary market was that they were not standardized.Answer: TRUETopic: Chapter 14.6 Secondary Mortgage MarketQuestion Status: Previous Edition15) Mortgage-backed securities have declined in popularity in recent years as institutional investors have sought higher returns in other markets.Answer: FALSETopic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: Previous Edition16) Mortgage-backed securities are marketable securities collateralized by a pool of mortgages.Answer: TRUETopic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: Previous Edition17) Fannie Mae and Freddie Mac together either own or insure the risk on nearly one-fourth of America's residential mortgages.Answer: FALSETopic: Chapter 14.4 Mortgage-Lending InstitutionsQuestion Status: Previous Edition18) A FICO score below 660 is considered good while a score above 720 is likely to cause problems in obtaining a loan.Answer: FALSETopic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition19) Subprime loans are those made to borrowers who do not qualify for loans at the usual market rate of interest because of a poor credit rating or because the loan is larger than justified by their income.Answer: TRUETopic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: Previous Edition14.3 Essay1) How has the modern mortgage market changed over recent years?Topic: Chapter 14.1 What Are Mortgages?Question Status: Previous Edition2) Explain the features of mortgage loans that are designed to reduce the likelihood of default.Topic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition3) What are points? What is their purpose?Topic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition4) How does an amortizing mortgage loan differ from a balloon mortgage loan?Topic: Chapter 14.2 Characteristics of the Residential MortgageQuestion Status: Previous Edition5) Evaluate the advantages and disadvantages, from both the lender's and borrower's perspectives, of fixed-rate and adjustable-rate mortgages.Topic: Chapter 14.3 Types of MortgagesQuestion Status: Previous Edition6) Why has the online lending market developed in recent years and what are the advantages and disadvantages of this development?Topic: Chapter 14.4 Mortgage-Lending InstitutionsQuestion Status: Previous Edition7) Why may Fannie Mae and Freddie Mac pose a threat to the health of the financial system?Topic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: Previous Edition8) What are mortgage-backed securities, why were they developed, whattypes of mortgage-backed securities are there, and how do they work?Topic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: Previous Edition9) What are the benefits and side effects of securitized mortgages?Topic: Chapter 14.7 Securitization of MortgagesQuestion Status: Previous Edition10) Discuss the pros and cons of a subprime market for residential mortgages in the U.S.Topic: Chapter 14.8 What Is a Mortgage-Backed Security?Question Status: New Question。
Journal of Artificial Intelligence Research20(2003)195-238Submitted6/03;published12/03 Taming Numbers and Durations in the Model CheckingIntegrated Planning SystemStefan Edelkamp stefan.edelkamp@cs.uni-dortmund.de Baroper Straße301Fachbereich Informatik,GB IVUniversit¨a t Dortmund44221DortmundGermanyAbstractThe Model Checking Integrated Planning System(MIPS)is a temporal least commit-ment heuristic search planner based on aflexible object-oriented workbench architecture.Its design clearly separates explicit and symbolic directed exploration algorithms from the set of on-line and off-line computed estimates and associated data structures.MIPS has shown distinguished performance in the last two international planning com-petitions.In the last event the description language was extended from pure propositional planning to include numerical state variables,action durations,and plan quality objective functions.Plans were no longer sequences of actions but time-stamped schedules.As a participant of the fully automated track of the competition,MIPS has proven to bea general system;in each track and every benchmark domain it efficiently computed plansof remarkable quality.This article introduces and analyzes the most important algorithmic novelties that were necessary to tackle the new layers of expressiveness in the benchmark problems and to achieve a high level of performance.The extensions include critical path analysis of sequentially generated plans to generate corresponding optimal parallel plans.The linear time algorithm to compute the parallel plan bypasses known NP hardness results for partial ordering by scheduling plans with respect to the set of actions and the imposed precedence relations.The efficiency of this algorithm also allows us to improve the exploration guidance:for each encountered planning state the corresponding approximate sequential plan is scheduled.One major strength of MIPS is its static analysis phase that grounds and simplifies parameterized predicates,functions and operators,that infers knowledge to minimize the state description length,and that detects domain object symmetries.The latter aspect is analyzed in detail.MIPS has been developed to serve as a complete and optimal state space planner, with admissible estimates,exploration engines and branching cuts.In the competition version,however,certain performance compromises had to be made,includingfloating point arithmetic,weighted heuristic search exploration according to an inadmissible estimate and parameterized optimization.1.IntroductionPractical action planning and model checking appear to be closely related.The MIPS project targets the integration of model checking techniques into a domain-independent action planner.With the HSF-Spin experimental model checker(Edelkamp,Leue,&Lluch-Lafuente,2003)we are looking towards the integration of planning technology into anEdelkampexisting model checker.Additional synergies are exploited in the automated compilation of protocol software model checking problems into planner inputs(Edelkamp,2003).Model checking(Clarke,Grumberg,&Peled,1999)is the automated process to verify if a formal model of a system satisfies a specified temporal property or not.As an illustrative example,take an elevator control system together with a correctness property that requires an elevator to eventually stop on every call of a passenger or that guarantees that the door is closed,while the elevator is moving.Although the success in checking correctness is limited, model checkers have found many errors in current hardware and software designs.Models often consist of many concurrent sub-systems.Their combination is either synchronous,as often seen in hardware design verification,or asynchronous,as frequently given in commu-nication and security protocols,or in multi-threaded programming languages like Java.Model checking requires the exploration of very large state spaces containing all reach-able system states.This problem is known as the state explosion problem and occurs even when the sets of generated states is much smaller than the set of all reachable states.An error that shows a safety property violation,like a deadlock or a failed assertion, corresponds to one of a set of target nodes in the state space graph.Roughly speaking, something bad has occured.A liveness property violation refers to a(seeded)cycle in the graph.Roughly speaking,something good will never occur.For the case of the elevator example,eventually reaching a target state where a request button was pressed is a liveness property,while certifying closed doors refers to a safety property.The two main validation processes in model checking are explicit and symbolic search. In explicit-state model checking each state refers to afixed memory location and the state space graph is implicitly generated by successive expansions of state.In symbolic model checking(McMillan,1993;Clarke,McMillan,Dill,&Hwang,1992), (fixed-length)binary encodings of system states are used,so that each state can be rep-resented by its characteristic function representation.This function evaluates to true if and only if all Boolean state variables are assigned to bit values with respect to the binary encoding of the system state.Subsequently,the characteristic function is a conjunction of literals with a plain variable for a1in the encoding and a negated variable for a0.Sets of states are expressed as the disjunction of all individual characteristic functions.The unique symbolic representation of sets of states as Boolean formulae through binary decision diagrams(BDDs)(Bryant,1992)is often much smaller than the explicit one.BDDs are(ordered)read-once branching programs with nodes corresponding to variables,edges corresponding to variable outcomes,and each path corresponding to an assignment to the variables with the resulting evaluation at the leaves.One reason for the succinctness of BDDs is that directed acyclic graphs may express exponentially many paths.The transition relation is defined on two state variable sets.It evaluates to true,if and only if an operator exists that transforms a state into a valid successor.In some sense,BDDs exploit regularities of the state set and often appear well suited to regular hardware systems.In contrast,many software systems inherit a highly asynchronous and irregular structure,so that the use of BDDs with afixed variable ordering is generally notflexible enough.For symbolic exploration,a set of states is combined with the transition relation to compute the set of all possible successor states,i.e.the image.Starting with the initial state,iteration of image computations eventually explores the entire reachable state space.Taming Numbers and Durations in MIPSTo improve the efficiency of image computations,transition relations are often provided in partitioned form.The correspondence between action and model checking(Giunchiglia&Traverso,1999) can be roughly characterized as follows.Similar to model checkers,action planners im-plicitly generate large state spaces,and both exploration approaches are based on applying operators to the current state.States spaces in model checking and in planning problems are often modelled as Kripke structures,i.e.state space graphs with states labelled by propo-sitional predicates.The satisfaction of a specified property on the one side corresponds to a complete exploration,and an unsolvable problem on the other side.In this respect,the goal in action planning can be cast as an error with the corresponding trail interpreted as a plan.In the elevator example,the goal of a planning task is to reach a state,in which the doors are open and the elevator is moving.Action planning refers to safety properties only, since goal achievement in traditional and competition planning problems have not yet been extended with temporal properties.However,temporally extended goals are of increasing research interest(Kabanza,Barbeau,&St-Denis,1997;Pistore&Traverso,2001;Lago, Pistore,&Traverso,2002).In contrast to model checkers that perform either symbolic or explicit exploration,MIPS features both.Moreover,it combines symbolic and explicit search planning in the form of symbolic pattern databases(Edelkamp,2002b).The planner MIPS implements heuristic search algorithms like A*(Pearl,1985)and IDA*(Korf,1985)for exploration,which include state-to-goal approximation into the search process to rank the states to be expanded next. Heuristic search has brought considerable gains to both planning(Bonet&Geffner,2001; Refanidis&Vlahavas,2000;Hoffmann&Nebel,2001;Bertoli,Cimatti,&Roveri,2001a; Jensen,Bryant,&Veloso,2002;Feng&Hansen,2002)and model checking(Yang&Dill, 1998;Edelkamp et al.,2003;Groce&Visser,2002;Bloem,Ravi,&Somenzi,2000;Ruys, 2003).Including resource variables,like the fuel level of a vehicle or the distance between two different locations,as well as action duration are relatively new aspects for competitive planning(Fox&Long,2003).The input format PDDL2.1is not restricted to variables of finite domain,but also includes rational(floating-point)variables in both precondition and effects.Similar to a set of atoms described by a propositional predicate,a set of numerical quantities can be described by a set of parameters.Through the notation of PDDL2.1,we refer to parameterized numerical quantities as functions.For example,the fuel level might be parameterized by the vehicle that is present in the problem instancefile description.In the competition,domains were provided in different tracks according to different lay-ers of language expressiveness:i)pure propositional planning,ii)planning with numerical resources,iii)planning with numerical resources and constant action duration,iv)planning with numerical resources and variable action duration,and,in some cases,v)complex prob-lems usually combining time and numbers in more interesting ways.MIPS competed as a fully automated system and performed remarkably well in allfive tracks;it solved a large number of problems and was the only fully automated planner that produced solutions in each track of every benchmark domain.In this paper the main algorithmic techniques for taming rational numbers,objective functions,and action duration are described.The article is structured as follows.First, we review the development of the MIPS system and assert its main contributions.ThenEdelkampwe address the object-oriented heuristic search framework of the system.Subsequently,we introduce some terminology that allows us to give a formal definition of the syntax and the semantics of a grounded mixed numerical and propositional planning problem instance. We then introduce the core contributions:critical path scheduling for concurrent plans, and efficient methods for detecting and using symmetry cuts.PERT scheduling produces optimal parallel plans in linear time given a sequence of operators and a precedence relation among them.The paper discusses pruning anomalies and the effect of different optimization criteria.We analyze the correctness and efficiency of symmetry detection in detail.The article closes with related work and concluding remarks.2.The Development of MIPSThe competition version of MIPS refers to initial work(Edelkamp&Reffel,1999a)in heuristic symbolic exploration of planning domains with theµcke model checker(Biere, 1997).This approach was effective in sample puzzle solving(Edelkamp&Reffel,1998)and in hardware verification problems(Reffel&Edelkamp,1999).For implementing a propositional planner,wefirst used our own BDD library called StaticBdd,in which large node tables are allocated prior to their use.During the imple-mentation process we changed the BDD representation mainly to improve performance for small planning problems.We selected the public domain c++BDD package Buddy(Lind-Nielsen,1999),which is moreflexible.The planning process was semi-automated(Edelkamp &Reffel,1999b);variable encodings were provided by hand,while the representations of all operators were established by enumerating all possible parameter instances.Once the state space encoding and action transition relation werefixed,exploration in the form of a symbolic breadth-first search of the state-space could be executed.At that time,we were not aware of any other work in BDD-based planning such as the work of Cimatti et al. (1997),which is likely thefirst link to planning via symbolic model checking.The team used the model checker(nu)SMV as the basis with an atom-to-variable planning state encoding scheme on top of it.Later on,we developed a parser and a static analyzer to automate the inference of state encodings,the generation of the transition relations,and the extraction of solution paths. In order to minimize the length of the state encoding,the new analyzer clustered atoms into groups(Edelkamp&Helmert,1999).As confirmed by other attempts(Weism¨u ller,1998), who started experimenting with PDDL specification inµcke,state minimization is in fact crucial.The simple encoding using one variable for each atom appears not to be competitive with respect to Graphplan-based(Blum&Furst,1995)and SAT-plan based planners(Kautz &Selman,1996).Subsequently,MIPS was thefirst fully automated planning system based on symbolic model checking technology that could deal with large domain descriptions.In the second international planning competition MIPS(Edelkamp&Helmert,2001) could handle the STRIPS(Fikes&Nilsson,1971)subset of the PDDL language(McDer-mott,2000)and some additional features from ADL(Pednault,1989),namely negative preconditions and(universal)conditional effects.MIPS was one offive planning systems to be awarded for“Distinguished Performance”in the fully automated track.The competition version(Edelkamp&Helmert,2000)already included explicit heuristic search algorithms based on a bit-vector state representation and the relaxed planning heuristic(RPH)(Hoff-Taming Numbers and Durations in MIPSmann&Nebel,2001)as well as symbolic heuristic search based on the HSP heuristic(Bonet &Geffner,2001)and a one-to-one atom derivative of RPH.In the competition,we used breadth-first bi-directional symbolic search whenever the single state heuristic search engine got stuck in its exploration.In between the planning competitions,explicit(Edelkamp,2001c)and symbolic pattern databases(Edelkamp,2002b)were proposed as off-line estimators for completely explored problem abstractions.Roughly speaking,pattern database abstractions slice the state vec-tor offluents into pieces and adjust the operators accordingly.The completely explored subspaces then serve as admissible estimates for the overall search and are competitive with the relaxed planning heuristic in several benchmark domains.For the third planning competition new levels of the planning domain description lan-guage(PDDL)were designed.Level1considers pure propositional planning.Level2also includes numerical resources and objective functions to be minimized.Level3additionally allows the specification of actions with durations.Consequently,MIPS has been extended to cope with these new forms of expressiveness.First results of MIPS in planning PDDL2.1problems are presented in(Edelkamp, 2001b).The preliminary treatment illustrates the parsing process in two simple bench-mark domains.Moreover,propositional heuristics and manual branching cuts were applied to accelerate sequential plan generation.This work was extended in(Edelkamp,2002a), where we presented two approximate exploration techniques to bound and tofix numerical domains,first results on symmetry detection based on fact groups,critical path scheduling, an any-time wrapper to produce optimal plans,and a numerical extension to RPH.3.Architecture of MIPSFigure1shows the main components of MIPS and the dataflow from the input definition of the domain and the problem instance to the resulting temporal plan in the output.As shown shaded in light gray,MIPS is divided into four parts:pre-compilation,heuristics, search algorithms,and post-compilation(scheduling).Henceforth,the planning process will be coarsely grouped into three stages,pre-compilation,heuristic search planning,and the construction of temporal plans.The problem and domain descriptionfiles are fed into the system,analyzed and grounded.Thisfixes the state space problem to be solved.The intermediate result is implicit,but can be saved in afile for use by other planners and model checkers.The basics of pre-compilation are covered in Section3.2.The next stage defines the planning process.The object-oriented workbench design of the planner allows different heuristic estimates to be combined with different search strategies and access data structures.Possible choices are listed in Sections3.3and 3.4. Temporal planning is based on(PERT)scheduling.This issue of rearranging sequential (relaxed)plans is addressed in detail in Section4.3.The planning system was developed in the spirit of the heuristic search framework, HSF for short(Edelkamp,1999),which allows attachment of newly implemented problem (puzzle)domains to an already compiled system.Similar to the approach that we took in model checking within HSF-Spin,we kept the extensible and general design.In fact we characterized both action planning and protocol validation as single-agent challenges. In contrast to the model checking approach,for planning we devised a hierarchy of systemEdelkampFigure1:Architecture of MIPS.states:the implementation for numerical states is a derived class of the one for propositional states.Similarily,the heuristic search algorithms are all based on an abstract search class.The main procedures that have to be provided to the search algorithm are a state expansion procedure,and a heuristic search evaluation function,both located in one of the hierarchi-cally organized heuristic estimator classes.In this sense,algorithms in MIPS are general node expanding schemes that can be adapted to very different problems.Additional data structures for the horizon list Open and the visited list Closed are constructed as parame-ters of the appropriate search algorithms.As a result,the implementations of the heuristic search algorithms and the associated data structures in the planner MIPS almost match those in our model checker.3.1Example ProblemThe running example for the paper is an instance of a rather simple PDDL2.1problem in Zeno-Travel.It is illustrated in Figure2.The initial configuration is drawn to the left ofTaming Numbers and Durations in MIPSFigure2:An instance for the Zeno-Travel domain with initial state(left)and goal state(s) (right).thefigure and the goal configuration to its right.Some global and local numeric variable assignments are not shown.Figures3and4provide the domain and problem specifications1.The instance asks for a temporal plan tofly passengers(dan,scott,and ernie)located somewhere on a small map(including the cities city-a,city-b,city-c,and city-d)with an aircraft(plane) to their respective target destinations.Boarding and debarking take a constant amount of time.The plane has afixed fuel capacity.Fuel and time are consumed according to the distances between the cities and the travel speed.Fuel can be restored by refueling the aircraft.As a result,the total amount of fuel is also maintained as a numerical quantity.3.2PrecompilerThe static analyzer takes the domain and problem instance as an input,grounds its proposi-tional state information and infers different forms of planner independent static information. Parsing Our simple Lisp parser generates a tree of Lisp entities.It reads the inputfiles and recognizes the domain and problem name.To cope with variable typing,we tem-porarily assert constant typed predicates to be removed together with other constant predicates in a further pre-compilation step.Thereby,we infer a type hierarchy and an associated mapping of objects to types.Indexing Based on the number of counted objects,indices for the grounded predicates, functions and actions are devised.Since in our example problem we have eight ob-jects and the predicates at and in have two parameters,we reserve2·8·8=128 index positions.Similarly,the function distance consumes64indices,while fuel,Edelkamp(define(domain zeno-travel)(:requirements:durative-actions:typing:fluents)(:types aircraft person city)(:predicates(at?x-(either person aircraft)?c-city)(in?p-person?a-aircraft))(:functions(fuel?a-aircraft)(distance?c1-city?c2-city)(slow-speed?a-aircraft)(fast-speed?a-aircraft)(slow-burn?a-aircraft)(fast-burn?a-aircraft)(capacity?a-aircraft)(refuel-rate?a-aircraft)(total-fuel-used)(boarding-time)(debarking-time))(:durative-action board:parameters(?p-person?a-aircraft?c-city):duration(=?duration boarding-time):condition(and(at start(at?p?c))(over all(at?a?c))):effect(and(at start(not(at?p?c)))(at end(in?p?a))))[...](:durative-action zoom:parameters(?a-aircraft?c1?c2-city):duration(=?duration(/(distance?c1?c2)(fast-speed?a))):condition(and(at start(at?a?c1))(at start(>=(fuel?a)(*(distance?c1?c2)(fast-burn?a))))) :effect(and(at start(not(at?a?c1)))(at end(at?a?c2))(at end(increase total-fuel-used(*(distance?c1?c2)(fast-burn?a))))(at end(decrease(fuel?a)(*(distance?c1?c2)(fast-burn?a))))))(:durative-action refuel:parameters(?a-aircraft?c-city):duration(=?duration(/(-(capacity?a)(fuel?a))(refuel-rate?a))):condition(and(at start(<(fuel?a)(capacity?a)))(over all(at?a?c))):effect(at end(assign(fuel?a)(capacity?a)))))Figure3:Zeno-Travel domain description in PDDL2.1.slow-speed,fast-speed,slow-burn,fast-burn,capacity,and refuel-rate each reserve eight index positions.For the quantities total-fuel-used,boarding-time, debarking-time only a single fact identifier is st but not least we model duration as an additional quantity total-time.This special variable is the only one that is overwritten in the least commitment planning approach when scheduling plans as described in Section4.Taming Numbers and Durations in MIPS(define(problem zeno-travel-1)(:domain zeno-travel)(:objects plane-aircrafternie scott dan-personcity-a city-b city-c city-d-city)(:init(=total-fuel-used0)(=debarking-time20)(=boarding-time30) (=(distance city-a city-b)600)(=(distance city-b city-a)600)(=(distance city-b city-c)800)(=(distance city-c city-b)800)(=(distance city-a city-c)1000)(=(distance city-c city-a)1000)(=(distance city-c city-d)1000)(=(distance city-d city-c)1000)(=(fast-speed plane)(/60060))(=(slow-speed plane)(/40060))(=(fuel plane)750)(=(capacity plane)750)(=(fast-burn plane)(/12))(=(slow-burn plane)(/13))(=(refuel-rate plane)(/75060))(at plane city-a)(at scott city-a)(at dan city-c)(at ernie city-c)) (:goal(and(at dan city-a)(at ernie city-d)(at scott city-d)))(:metric minimize total-time))Figure4:Zeno-Travel problem instance.Flattening Temporal Identifiers We interpret each action as an integral entity,so that all timed propositional and numerical preconditions can be merged.Similarly,all effects are merged,independent of time at which they happen.Invariant conditions like(over all(at?a?c))in the action board are added into the precondition set.We discuss the rationale for this step in Section4.1.Grounding Propositions Fact-space exploration is a relaxed enumeration of the planning problem to determine a superset of all reachable facts.Algorithmically,a FIFO fact queue is compiled.Successively extracted facts at the front of the queue are matched to the operators.Each time all preconditions of an operator are fulfilled,the resulting atoms according to the positive effect(add)list are determined and enqueued.This allows us to separate offconstant facts fromfluents,since only the latter are reached by exploration.Clustering Atoms For a concise encoding of the propositional part we separatefluents into groups,so that each state in the planning space can be expressed as a conjunc-tion of(possibly trivial)facts drawn from each fact group(Edelkamp&Helmert, 1999).More precisely,let#p i(o1,...,o i−1,o i+1,...,o n)be the number of objects o i for which the fact(p o1...o n)is true.We establish a single-valued invariant at i if#p i(o1,...,o i−1,o i+1,...,o n)=1.To allow for a better encoding,some predicates like at and in are merged.In the example,three groups determine the unique position of the persons(one offive)and one group determines the position of the plane(one of four).Therefore,3·⌈log5⌉+1·⌈log4⌉=11bits suffice to encode the total of19fluents.EdelkampGrounding Actions Fact-space exploration also determines all grounded operators.Once all preconditions are met and grounded,the symbolic effect lists are instantiated.In our case we determine98instantiated operators,which,by some further simplifications that eliminate duplicates and trivial operators(no-ops),are reduced to43. Grounding Functions Simultanous to fact space exploration of the propositional part of the problem,all heads of the numerical formulae in the effect lists are grounded.In the example case only three instantiated formulae arefluent(vary with time):(fuel plane)with initial value750as well as total-fuel-used and total-time both ini-tialized with zero.All other numerical predicates are in fact constants that can be substituted in the formula-bodies.In the example,the effect in(board dan city-a) reduces to(increase(total-time)30),while(zoom plane city-a city-b)has the numerical effects(increase(total-time)150),(increase(total-fuel-used) 300)),and(decrease(fuel plane)300).Refuelling,however,does not reduce toa single rational number,for example the effects in(refuel plane city-a)onlysimplify to(increase(total-time)(/(-(750(fuel plane))/12.5)))and (assign(fuel plane)750).To evaluate the former assignment especially for a for-ward chaining planner,the variable(total-time)has to be instantiated on-the-fly.This is due to the fact that the value of the quantity(fuel plane)is not constant and itself changes over time.Symmetry Detection Regularities of the planning problem with respect to the transpo-sition of domain objects is partially determined in the static analyzer and is addressed in detail in Section5.The intermediate textual format of the static analyzer in annotated grounded PDDL-like representation serves as an interface for other planners or model checkers,and as an additional resource for plan visualization.Figures5and6show parts of the intermediate representation as inferred in the Zeno-Travel example.3.3HeuristicsMIPS incorporates the following heuristic estimates.Relaxed planning heuristic(RPH)Approximation of the number of planning steps needed to solve the propositional planning problem with all delete effects removed(Hoff-mann&Nebel,2001).The heuristic is constructive,that is it returns the set of operators that appear in the relaxed plan.Numerical relaxed planning heuristic(numerical RPH)Our extension to RPH to deal with with numbers is a combined propositional and numerical approximation scheme allowing multiple operator application.Pattern database heuristic(explicit PDB)Different planning space abstractions are found in a greedy manner,yielding a selection of pattern databases thatfit into main memory.In contrast to RPH,pattern database can be designed to be disjoint yielding an admissible estimate as needed for optimal planning in A*(Edelkamp,2001c).(define(grounded zeno-travel-zeno-travel-1)(:fluents(at dan city-a)(at dan city-b)(at dan city-c)(at dan city-d)(at ernie city-a)(at ernie city-b)(at ernie city-c)(at ernie city-d)(at plane city-a)(at plane city-b)(at plane city-c)(at plane city-d)(at scott city-a)(at scott city-b)(at scott city-c)(at scott city-d)(in dan plane)(in ernie plane)(in scott plane))(:variables(fuel plane)(total-fuel-used)(total-time))(:init(at dan city-c)(at ernie city-c)(at plane city-a)(at scott city-a)(=(fuel plane)750)(=(total-fuel-used)0)(=(total-time)0))(:goal(at dan city-a)(at ernie city-d)(at scott city-d))(:metric minimize(total-time))(:group dan(at dan city-a)(at dan city-b)(at dan city-c)(at dan city-d)(in dan plane))(:group ernie(at ernie city-a)(at ernie city-b)(at ernie city-c)(at ernie city-d)(in ernie plane))(:group plane(at plane city-a)(at plane city-b)(at plane city-c)(at plane city-d))(:group scott(at scott city-a)(at scott city-b)(at scott city-c)(at scott city-d)(in scott plane))Figure5:Grounded representation of Zeno-Travel domain.Symbolic pattern database heuristic(symbolic PDB)Symbolic PDBs apply to ex-plicit and symbolic heuristic search engines(Edelkamp,2002b).Due to the succinct BDD-representation of sets of states,symbolic PDBs are often orders of magnitudes larger than explicit ones.Scheduling relaxed plan heuristic(scheduling RPH)Critical-path analysis through scheduling guide the planfinding phase.Like RPH,which computes the length of the greedily extracted sequential plan,scheduling RPH also takes the relaxed sequence of operators into account,but searches for a suitable parallel arrangement,which in turn defines the estimator function.3.4Exploration AlgorithmsThe algorithm portfolio includes three main explicit heuristic search algorithms.A*The A*algorithm(Hart,Nilsson,&Raphael,1968)is a variant of Dijkstra’s single-source shortest path exploration scheme executed on a re-weighted state space graph.For lower bound heuristics,A*can be shown to generate optimal plans(Pearl,1985).Weighting the influence of the heuristic estimate may accelerate solutionfinding,but also affects optimality(Pohl,1977).。