Chapter 7
7.2 (i)
The coefficient of cigs in the first equation is negative 0.0044, which means if smoking one more cigarette, the birth weight of the baby would reduce 0.44%, holding the variables of faminc and parity fixed.
If cigs = 10 then = 0.0044*10= 大度读人0.044=4.4%, which means about a 4.4% lower birth weight.
(ii)
Holding the other factors in the first equation fixed, a white child predict to weigh higher 5.5%(=0.055=5.5%) than a nonwhite child.
What’s more, twhite =0.055/ 0.013 4.23, which is well above any commonly ud critical value. grapheneThus, the difference between white and nonwhite babies is also statistically significant.
(iii)
The coefficient of motheduc is about negative 0.003, which means if mother’s education years increa one more year, the birth weight of the baby would reduce 0.3%, holding the other factors in the cond equation fixed. What’s more, t motheduc =0.003/ 0.003=1, which is 汇佳小学below any commonly ud critical value. Thus, the difference of mother’s education years is not statistically significant.
(iv)
From the given information, I’m unable to compute the F statistic for joint significance of motheduc and fatheduc, the reason is that the two regressions u different ts of obrvations. The cond regression us fewer obrvations becau motheduc or fatheduc are missing for some obrvations. If we want to compute the F statistic, we would have to reestimate the first equation using the same obrvations which is ud to estimate the cond equation.
7.4 (i)
cucu
把中文翻译成韩语The coefficient of calculus is positive 4.41, which means if a student has learned about binary variable in the calculus, the percentages of his or her cour in the total scores will increa by 441%. This evaluation is not reasonable.
(ii)
When we control the variable of msugpa, the performances in high school help to predict the scores of Microeconomics Theory.
(iii)
When we control other variables, the degree of parents doesn’t help to predict the scores of Microeconomics Theory.
7.9 (i)
From the equation y= , we know that u=0 and d=1, so we can get f1(z)= .
(ii)
Set f0(z*)= f1(z*), so we can get , compute and get , considering , so z*=-. Only when is negative, z* is positive, which means and must have opposite signs.
(iii)
Using part (ii), becau of and , the z*=totcoll==0.357/0.030=11.9 years.
(iv)
From the part (iii), the estimated years of college where women catch up to men is about 11.9 years. However, the average years of college are about 4 years. So it is not real to exceed the men’s wages with the four years’ college for women.
C7.3 (i)重庆科技学院是几本
H0: = 0. Becau the catcher is as the ba group. Using the data in MLB1.RAW gives = 0.254, ()=0.131. So the t statistic is about 1.93, which gives a p-value aboutauditor 0.054. Therefore, we would reject H0 at just about the 5% significance level. The estimated salary difference between catchers and outfielders is about 28.92% (100*[exp(0.254) – 1]).
(ii)
H0: = 0; = 0; = 0; = 0; = 0.
F=（SSR-SSR1-SSR2）/(SSR1+SSR2)*(N-2k-2)/(k+1)
=
The F statistic, with 5 and 339 df, is about 1.78, and its 徐星海p-value is about .117. Thus, we cannot reject H0 at the 10% level.
(iii)
(iii) Parts (i) and (ii) are roughly consistent. The evidence against the joint null in part (ii) is weaker becau we are testing, along with the marginally significant catcher, veral other insignificant variables (especially thrdba and shrtstop, which has absolute t statistics well below one).
C7.12 (i)
girl friendFor men, the parate fractions rated that are classified as having above average looks is about 0.29. For women, it is about 0.3303.
There are more people rated as having above average, becau the proportion of people rated as having below average looks is just about 0.123, but the proportion of people rated as having above average looks is higher at about 0.304.
windows live是什么(ii)
From the part (i), we know that the difference having above average looks between male and female is about 0.0403(=0.3303-0.29), which means the the percent rated as having above average looks is higher for women than for men. A way to test whether the population fractions of above average looking women and men are the same is to run a simple linear probability model about abvavg on female. From the upper data, the t statistic is about 1.48 with two-sided p-value = 0.140 or with the one-sided p-value=0.070. Therefore, for the hypothesis that the population fractions are the same, there is not strong evidence against.
(iii)
For the women, the results of the regression is
Log(wage)=1.3088-0.1376*belavg+0.0336*abvavg;
n=436; R²=0.0105.
The coefficient of belavg is about negative 0.1376, which means that if the women is below the average looking, her wage will reduce 13.76% less than the women with average looks.
