Problem Set 5


Problems from Textbook

4.20

Additional Problem

A recent issue of Newsweek featured an article on California's proposal to stop using the SAT in College admissions. The article reported the differences in average scores by gender and ethnic group. However, Newsweek fails to report standard deviations as well as the number of observations from which the averages were constructed. This makes interpretation of the differences difficult. The source of Newsweek's information is the College Board, which reports the averages and standard deviations as well the number of students who took the test.

  1. Is the 3 point difference between the average SAT verbal scores of white males and white females on statistically significant?
  2. Is the difference between the population mean scores of white males and white females significantly greater than 2 points?
  3. Is the variance in scores among black students significantly greater than the variance of scores among white students?

Computer Exercise

The dataset returns.txt contains annual data for following variables:

Load the data returns.txt to stataquest and do the following exercises:
  1. Excess return on a risky portfolio is the difference between the portfolio return and the return on a riskless bond - e.g. U.S. treasury bill. Calculate excess returns on S&P500 and DJI.
  2. Test the hypothesis that excess returns on S&P 500 are different from zero (use 5% significance).
  3. Test the hypothesis that excess returns on S&P 500 are different from excess returns on DJI (use 5% significance).
  4. Test if the variance of S&P 500 returns is higher than the variance of DJI returns (use 5% significance).