y = beta0 + beta1*x + u
beta1=sum [(xi-x)(yi-y)] / sum [(xi-x)^2]
(a) Is the estimator beta1 a random variable? Why or why not?
(b) Specify conditions under which beta1 is unbiased.
(c) Now assume that E (u|x) = 1 instead of 0. Is beta1 still unbiased?
linear regression question, plz plz help
2008-02-06 12:49 pm
回答 (1)
2008-02-06 7:36 pm
✔ 最佳答案
(a)β1 is the random variable, because x and y are random variable
when use different values of (xi and yi) from each experiment, we get different answer
(b)
if E(ui)=0 for all i, then beta1 is unbiased
(c)
No, from the answer of (b), β1 is not unbiased
收錄日期: 2021-04-26 13:05:21
原文連結 [永久失效]:
https://hk.answers.yahoo.com/question/index?qid=20080206000051KK00388