计量经济学-实验三

浙江财经大学
实验（实训）报告
项目名称logistic or probit model
所属课程名称计量经济学
项目类型验证性实验
实验(实训)日期15年04月日
班级
学号
姓名
指导教师
浙江财经大学教务处制
实验三报告
Logistic model，Probit model（验证性实验）
实验类型：验证性实验
实验目的：
当被解释变量是虚拟变量时，学会用Logistic model或Probit model进行估计，掌握似然比（LR）检验，学会解释模型的估计值。

实验内容：
Logistic model或Probit model的估计
实验要求：
掌握Logistic model或Probit model的估计，按具体的题目要求完成实验报告,并及时上传到给定的FTP！
实验题目：
[abstracted from << Introductory Economerics >>chapter17 C17.8]
The file JTRAIN2.dta ontains data on a job training experiment
for a group of men. Men could enter the program starting in January 1976 up through about mid-1977.The program ended in December 1977.The idea is to test whether participation in the job training program had an effect on unemployment probabilities and earnings in 1978. 就业培训是否对失业率及收益有影响
(i)The variable train is the job training indictor. How many men in the example participated in the job training program? What was the highest number of months a man actually participated in the program? (consider the variable mosinex).
(ii)Run a linear regression of train on several demographic and pretraining variables:unem74,unem75,age,educ,black,hisp,and married. Are these variables jointly significant at the 5% level?
(iii)Estimate a probit version of the linear model in part(ii).Compute the likelihood ratio test for joint significance of all variables .What do you conclude?
(iv) Run a simple regression of unem78 on train and report the results in equation form. What is the estimated effect of participating in the job training program on the probability of being unemployed in 1978? Is it statistically significant?
(v)Run a probit of unem78 on train .Does it make sense to compare the probit coefficient on train with the coefficient obtained from the linear model in part(v)? (vi)Find the fitted probabilities from parts(v) and (vi).Explain why they are identical. Which approach would you use to measure the effect and statistical significance of the job training program?
(vii)Add all of the variables from part(ii) as additional controls to the models from parts(v) and (vi).Are the fitted probabilities now identical? What is the correlation between them?
实验题目分析报告：
(i)
sum train if train==1
445人中有185人参加就业培训计划
sum mosinex
实验中时间最长的为24个月
(ii)
reg train unem74 unem75 age educ black hisp married
F（7，437）=1.43.p=0.1915，5%的置信水平上联合显著
(iii)
probit train unem74 unem75 age educ black hisp married
P(train = 1|x)=Φ(β0+ β1unem74 + β2unem75 + β3age + β4educ + β5black + β6hisp + β7married)
LR chi2（7）=10.18，p=0.1785，和第二题中LPM 获得的近似。

(iv) reg unem78 train
unem= 0.35 - 0.11train
（0.028） （0.044）
n=445，2R =0.0139 参加在职培训的在1987年失业率下降了0.111，这是很大的影响，没有参加培训的失业率为0.354，培训将失业率降低至0.243，这个差异在1%的双侧检验下具有显著的统计意义。

(v) probit unem78 train
)321.0375.0()|178ˆun (train train m e
P --Φ== (0.080) (0.128)
与题目四模型中的系数比较无意义，但两个模型的t 统计量相同。

(vi)
qui reg unem78 train
predict lhat
(option xb assumed; fitted values)
tabulate lhat
qui probit unem78 train
predict phat
(option pr assumed; Pr(unem78))
tabulate phat
(vii)
qui reg unem78 train unem74 unem75 age educ black hisp married
predict l2hat
(option xb assumed; fitted values)
qui probit unem78 train unem74 unem75 age educ black hisp married
predict p2hat
(option pr assumed; Pr(unem78))
corr p2hat l2hat
拟合的值已不再完全相同，因为模型不饱和，解释变量不是详尽的、互相排斥的一组虚拟变量。

但由于其他解释变量是微不足道的，而且都是高度相关拟合的值，拟合的值不完全相同，他们之间有0.9932的相关。