DOC PREVIEW
UCSB ECON 240a - Midterm Key

This preview shows page 1-2 out of 6 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 6 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 6 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 6 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Oct. 31, 2002 Midterm Key L. PhillipsEcon 240A-1Answer all five queations.1. (15 points) An investor believes that on a day when the Dow Jones Industrial Average (DJIA) increases, the probability that the NASDAQ also increases is 77%. If the investor believes that there is a 60% probability that the DJIA will increase tomorrow, what is the probability that the NASDAQ will increase as well? Answer: Prob(NASDAQ up/DJIA up) = 0.77Prob(DJIA up tomorrow) = 0.60Prob (NASDAQ up) = Prob(NASDAQ up/DJIA up)* Prob(DJIA up tomorrow)= 0.77*0.60 = 0.4622. (15 points) The boss assigns you to choose an internet service provider (ISP). You want an ISP that is large enough so that a customer (caller) seldom receives a busy signal. With the ISP you choose, a customer (caller) encounters a busy signal only 8%of the time. Your office makes 60 calls per week.a. What is the probability that callers from your office do not encounter any busysignals in a week? Note: Full credit for the correct numerical expression for the probability. You do not have to calculate the numerical value of this expression in part a.Answer: each call is a Bernoulli event with a success being no busy signal with probability 0.92 and a failure being a busy signal with probability 0.08. Using the binomial distribution, the probability of 60 successes in 60 calls is:P(k=60) = 60!/(60!*0!)(0.92)60(0.08)0 = 1*)(0.92)60*1 = (0.92)60This is a satisfactory answer for this exam. LnP(k=60) = 60*ln(0.92) =-5.0028965So exp(5.002895 = 0.0067 = Prob(k=60)01.0Oct. 31, 2002 Midterm Key L. PhillipsEcon 240A-2b. Suppose you want to approximate an answer for this probability. Provide a numerical approximate value for the probability that callers from your office do not encounter a busy signal in a given week.Answer: Using the normal approximation, the E(k) = n*p = 60*0.92 = 55.2.The VAR(k) = n*p*(1-p) = 60*(0.92)*(0.08) = 4.416. The standard deviation is the square root of 4.41 which equals 2.1. The probability of getting less than 60 successes would be the probability of falling below a standardized z of:Z = (60-55.2)/2.1 = 2.29. From Table 3 in Appendix B, the area in the upper tail of the normal distribution for z = 2.29 is 0.011, so the chances that no one in your office gets a busy signal are pretty low. c. Is this approximation pretty good? Explain.Answer: yes the criteria for the approximation are n*p5and n*(1-p)5. The first is 60*0.92 and is clearly satisfied, the second is 60*0.08 = 4.8 and is close.3. (15 points) Americans tend to eat too much fast food. A doctor claims that the averageCalifornian is more than 20 pounds overweight. To check his claim, you take a random sample of 20 Californians and weigh them. The difference between their observed weight and their ideal weight follows: 16, 23, 18, 41, 22, 18, 23, 19, 22, 15, 18, 35, 16, 15, 17, 19, 23, 15, 16, 26. The mean of this sample is 20.85, and the standard deviation of this sample is 6.76.a. From this sample data, do you think the doctor's claim is true?Answer: The null hypothesis is that the population mean is equal to 20, and the alternative hypothesis is that the population mean is >20. Since the population mean is not known, use Student’s t-distribution. The t-statistic is:t = nsxEx /)]([  = [20.85 – 20]/(6.720) = 0.85/(6.76/4.47) = 0.85/1.51=0.56, where E()x= the population mean under the null hypothesis, i.e.Oct. 31, 2002 Midterm Key L. PhillipsEcon 240A-320. So do not reject the null. For 19 degrees of freedom, i.e n-1, at the 5% level the critical t-statistic is 1.73, so the calculated t-statistic is well below the critical t.for a 5% type I error.4. (15 points) The following graph plots the UC budget, the component funded by the state, against California Personal Income, both in billions of nominal dollars. A linear trendline has been fitted to the data.a. How would you describe the goodness of fit?Answer: It is fairly good in that R2 is 95%, but there are some sizeable errors in the 80’s and 90’s, before and after the recession of 1991.b. Would you use only this trendline to predict next year's UC budget?Answer: It would be a good idea to calculate several forecasts, for example one from problem 5 below, as well.c. From the information provided about past experience, if California Personal Income goes up by a 100 billion next year, how much would you expect the UC budget to increase?Answer: From the slope of 0.0027, about 0.27 billion or 270 millionOct. 31, 2002 Midterm Key L. PhillipsEcon 240A-4Figure 4.1 UC Budget, General Fund Component Vs. CA Personal Income, both in Billions of Nominal Dollars, 1968-69 through 2002-20035. (15 points) In the figure below, the UC Budget, General Fund Component, is plotted against California Personal Income, both in billions of nominal dollars, from fiscal year 1968-69 through 2002-03, on a log-log scale. In the table that follows, the resultsof regressing the natural logarithm of the UC Budget against the natural logarithm of California Personal Income follows.a. Is this regression statistically significant? Explain.Answer: yes, the F-statistic of 1598 is highly significant.b. Interpret the estimated slope coefficient.Answer: it is the elasticity of the UC Budget to CA personal Income, indicating for every 10% increase in CA personal Income, the UC Budget goes up 8.78%.c. Is this slope coefficient significantly different from zero?UC Budget, General Fund Component Vs. CA Personal Income, Both in Billions of Nominal $, 1968-69 through 2002-2003y = 0.0027x + 0.1972R2 = 0.951300.511.522.533.540 200 400 600 800 1000 1200CA Personal Income UC BudgetOct. 31, 2002 Midterm Key L. PhillipsEcon 240A-5Answer: Yes, the t-statistic from Table 5.1 of 40 is highly significant.d. Is this slope significantly less than one?Answer: the null hypothesis is that the slope is one and the alternative hypothesis is that the slope is less than one. The t-statistic is:)ˆ(/)ˆ(ˆ[E] = (0.878-1.0)/0.0219 = -0.122/0.0219 = -5.57 where the expected value of the estimated slope is 1 under the null hypothesis. So the elasticity is significantly less than one.From Table 4 in Appendix B, for 33 = n-2 degrees of freedom, the critical t-statistic is –1.70, since the t-distribution is symmetric.e. Which model appears to fit better, the linear one or the proportional(power function) one? Answer: the proportional has a higher R2, but they are comparable.Figure 5.1 UC Budget, General Fund Component Vs. CA Personal Income, both in


View Full Document

UCSB ECON 240a - Midterm Key

Documents in this Course
Final

Final

8 pages

power_16

power_16

64 pages

final

final

8 pages

power_16

power_16

64 pages

Power One

Power One

63 pages

midterm

midterm

6 pages

power_16

power_16

39 pages

Lab #9

Lab #9

7 pages

Power 5

Power 5

59 pages

Final

Final

13 pages

Final

Final

11 pages

Midterm

Midterm

8 pages

Movies

Movies

28 pages

power_12

power_12

53 pages

midterm

midterm

4 pages

-problems

-problems

36 pages

lecture_7

lecture_7

10 pages

final

final

5 pages

power_4

power_4

44 pages

power_15

power_15

52 pages

group_5

group_5

21 pages

power_13

power_13

31 pages

power_11

power_11

44 pages

lecture_6

lecture_6

12 pages

power_11

power_11

42 pages

lecture_8

lecture_8

11 pages

midterm

midterm

9 pages

power_17

power_17

13 pages

power_14

power_14

55 pages

Final

Final

13 pages

Power One

Power One

53 pages

Summary

Summary

54 pages

Midterm

Midterm

6 pages

Lab #7

Lab #7

5 pages

powe 14

powe 14

32 pages

Lab #7

Lab #7

5 pages

Midterm

Midterm

8 pages

Power 17

Power 17

13 pages

Midterm

Midterm

6 pages

Lab Five

Lab Five

30 pages

power_16

power_16

64 pages

power_15

power_15

52 pages

Power One

Power One

64 pages

Final

Final

14 pages

Load more
Download Midterm Key
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Midterm Key and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Midterm Key 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?