課程名稱︰計量二
課程性質︰選修
課程教師︰劉錦添
開課學院：
開課系所︰經濟系
考試日期（年月日）︰2014.6.13
考試時限（分鐘）：180 mins
是否需發放獎勵金：yes
（如未明確表示，則不予發放）
試題 :
1. The data in FERTIL2.RAW includes, for women in Botswana during 1988,
information
on number of children, years of education, age, and religious and economic
status
variables.
(a) Estimate this model by OLS
children = β0 + β1 educ + β2 age + β3 age^2 + u
and interpret the estimates. In particular, holding age fixed, what is the
estimated effect of another year of education on fertility? If 100 women
receive another year of education, how many fewer children are they
expected to have?
(b) Frsthalf is a dummy variable equal to one if the woman was born during
the first six months of the year. Assuming that frsthalf is uncorrelated
with the error term from part (a), show that frsthalf is a reasonable
IV candidate for educ. (Hint: You need to do a regression.)
(c) Estimate the model from part (a) by using frsthalf as an IV for educ.
Compare the estimated effect of education with the OLS estimate from
part (a).
(d) Add the binary variables electric, tv, and bicycle to the model and
assume these are exogenous. Estimate the equation by OLS and 2SLS
and compare the estimated coefficients on educ. Interpret the coefficient
on tv and explain why television ownership has a negative effect
on fertility.
2. Bernard and Busse use the Olympic Games data inolympics.dat to examine the
share
of medals won by countries. The total number of medals awarded in 1988 was
738; in
1992 there were 815 medals awarded, and in 1996, 842 medals were awarded.
Using
these totals, compute the share of medals (SHARE) won by each country in each
of these years.
(a) Construct a histogram for the variable SHARE. What do you observe? What
percent of the observations are zero?
(b) Estimate a least squares regression explaining SHARE as a function of the
logarithms of population and real GDP, and the variables HOST and SOVIET.
(i) Discuss the estimation results.
(ii) Plot the residuals against ln(GDP). Do they appear random?
(iii) Use your computer software to compute the skewness and kurtosis values
of the residuals. How do these values compare to those for the normal
distribution, which has skewness of zero and kurtosis of three?
(c) In 2000, the GDP (in 1995 US $) of Australia was 3.22224E + 11, and that
of
Canada was 6.41256E + 11. The Australian population in 2000 was 19.071
million, and that of Canada was 30.689 million. Predict the share of medals
won
by Canada and Australia based on the estimates in part (b). Note that the
2,000
games were held in Sydney, Australia. In 2000, Australia won 58 medals and
Canada won 14 out of the 929 medals awarded. How close were your predicted
shares?
(d) Estimate the model described in (b) using Tobit. Compare the parameter
estimates to those in (b).
3. Is there a formula that can predict the outcome the Oscar Award for the
best Picture? The file oscar.dta contains informations on the nominees for
Best Picture since 1984. The variable are YEAR, TITLE, a dummy variable
WINNER indicating the Oscar, the total number of NOMINATIONS, the total
number of Golden Globe (GGLOBES) won, and a dummy variable indicating whether
or not the film was a comedy (COMEDY).
(a) Estimate a Probit model to predict the Oscar winning using NOMINATIONS
and GGLOBES. Are the coefficients siginicant and of anticipated signs?
(b) Calculate the marginal effect of an additional NOMINATION on the
probability of winning a Oscar.
(c) Calculate the marginal effect of an additional GGLOBES on the probability
of winning a Oscar.
(d) Preduct the probability of winning for each film up through 2003. Using
the rule that the highest predicted probability predicts the winner, compute
the percentage of correct predictions by this model.