Question 1

The flaw of "not having a representative sample" means that

a. the number of interview subjects is too small to be meaningful.

b. the characteristics of the pool of respondents are significantly different from the characteristics of the entire population of interest.

c. the interview subjects have not agreed to be studied.

d. the interview subjects know that they are part of an experiment and thus behave differently than they would in a normal setting.

Question 2

"Response bias" occurs when

a. interview subjects do not know what their future actions will be.

b. interview subjects provide answers that differ from their actual actions.

c. the interviewer prompts subjects to give particular responses for the sake of the study.

Question 3

You estimate a demand function for widgets and find that Q=203P+0.04M1.8PrQ=203P+0.04M1.8Pr, where Pr is the price of a related good, the wodget. These estimated coefficients support the law of demand for widgets.

a. True

b. False

Question 4

You estimate a demand function for widgets and find that Q=203P+0.04M1.8PrQ=203P+0.04M1.8Pr, where Pr is the price of a related good, the wodget. The estimated coefficients suggest that widgets are an inferior good.

a. True

b. False

Question 5

You estimate a demand function for widgets and find that Q=203P+0.04M1.8PrQ=203P+0.04M1.8Pr, where Pr is the price of a related good, the wodget. The estimated coefficients suggest that widgets and wodgets are substitutes.

a. True

b. False

Question 6

A simple time-series model is estimated where widget demand Q=200+2.1tQ=200+2.1t, where t is the year beginning in 2001 (t = 1 in 2001, t = 2 in 2002, etc.). These results suggest that widget demand is expected to increase year-after-year.

a. True

b. False

Question 7

A simple time-series model is estimated where widget demand Q=200+2.1tQ=200+2.1t, where t is the year beginning in 2001 (t = 1 in 2001, t = 2 in 2002, etc.). The estimated demand for widgets in 2002 is ........ widgets.

Question 8

A simple time-series model is estimated where wodget demand Q=100+3.4tQ=100+3.4t, where t is the year beginning in 2001 (t = 1 in 2001, t = 2 in 2002, etc.); however, the coefficient for t is statistically insignificant. You would conclude that

a. wodget demand falls over time, since the insignificant t coefficient means you reverse the sign.

b. wodget demand is independent of time; i.e., doesn't change from year-to-year.

c. wodget demand predictably increases year-after-year.

Question 9

If you were to estimate an equation on monthly data where total employment (# of employees) at H&R Block, CPA, and tax firms was the dependent variable, you'd likely want to include a dummy variable for

a. time periods during the tax season (Feb. 1 through April 15).

b. the fourth quarter of the year.

c. summer months (June-Aug).

Question 10

If you collected monthly data on housing construction and wanted to estimate new housing starts, you would want to include dummy variables

a. for warm-weather months, as well as for the recent recession, since the housing market was strongly affected.

b. both for recessions and expansions to capture all phases of the business cycle.

c. for the Christmas season as well as for July.

d. both for warm-weather as well as for cold-weather months, to guarantee that all weather-related factors are accounted for.

Question 11

Assume you have daily retail sales data and hypothesize that sales are always higher on weekends than during the work week. You should include a dummy variable

a. for Saturday ( =1 on Saturday, =0 otherwise) and another dummy variable for Sunday ( =1 on Sunday, =0 otherwise).

b. for "weekend" ( =1 on Saturday and Sunday, =0 otherwise), and another dummy variable for "work week" (=1 for Monday-Friday, =0 otherwise).

c. for "weekend" which equals one on Saturday and Sunday and zero otherwise.