利用bootstrap进行显著性检验

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

运用Bootstrap的方法，从所有残差项里重新取样，重复1000次，
请问这个命令在stata里怎么写呢？

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1 收藏0 回帖

关键词：Bootstrap Bootstra boots Trap boot

"help bootstrap"

jjjj6666 发表于 2014-2-26 00:22
"help bootstrap"

help了 但是没看懂，我想对残差进行bootstrap检验

what kind of test you want to do for the residual (?) variable?

jjjj6666 发表于 2014-2-26 21:48
what kind of test you want to do for the residual (?) variable?

F值检验

I don't know which command you use for f-test.  

Taking an example, if based on reg, to bootstrap F for testing the coefficients 100 times, we can do

bootstrap e(F), rep(100) : reg y x

jjjj6666 发表于 2014-2-28 04:29
I don't know which command you use for f-test.  

Taking an example, if based on reg, to bootstrap ...

谢谢  我试一下

jjjj6666 发表于 2014-2-28 04:29
I don't know which command you use for f-test.  

Taking an example, if based on reg, to bootstrap ...

多谢指教

jjjj6666 发表于 2014-2-28 04:29
I don't know which command you use for f-test.  

Taking an example, if based on reg, to bootstrap ...

有个问题请教您：单独做PPML回归的时候结果很好，单独通过bootstrap抽样并用reg回归的时候结果也挺好，但是用bootstrap抽样后再用ppml 回归，出现错误提示Number of regressors excluded to ensure that the estimates exist: 0Number of observations excluded: 0，这该如何处理？请您指教，多谢

jjjj6666 发表于 2014-2-28 04:29
I don't know which command you use for f-test.  

Taking an example, if based on reg, to bootstrap ...

有个问题请教您：单独做PPML回归的时候结果很好，单独通过bootstrap抽样并用reg回归的时候结果也挺好，但是用bootstrap抽样后再用ppml 回归，出现错误提示Number of regressors excluded to ensure that the estimates exist: 0Number of observations excluded: 0，这该如何处理？请您指教，多谢