lincom, test, reg命令的多元回归问题

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

刚学stata,遇到下面这题。请大家帮帮忙，真心求教！


educ = 0 + β1 sibs + β2 meduc + β3 feduc + u.
1. Test at 1% level if the three variables are individually significant against a two-sided
alternative.
2. Test at 1% level if the coefficient of feduc is statistically different from 0.5 against a
two-sided alternative by rewriting the model appropriately. Then, verify your result
by using the “lincom” command.
3. Build a confidence interval (C.I.) at 99% confidence level of the parameter on feduc.
Can you reject the null hypothesis at point 2 using this C.I.? Why?
4. Test at 1% level the overall significance of the model using the SSR version of the
F-statistic (you have to estimate both the restricted and the unrestricted model).
Then, verify your result by using the “test” command.

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：lincom 多元回归 test Est NCO different against verify

reg educ sibs mduc feduc, level(99).lincom feduc-0.5. test feduc=0.5???是这样？