Confidence intervals of linear multiple regression
조회 수: 3 (최근 30일)
이전 댓글 표시
Considering following linear multiple regression model:
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/570219/image.png)
where observations
, coefficents
and
is a white Gaussian noise term.
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/570224/image.png)
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/570229/image.png)
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/570234/image.png)
By the definition of the confidence interval:
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/570239/image.png)
where
is the critical value, and standard error
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/570244/image.png)
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/570249/image.png)
I understand that
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/570254/image.png)
My question is that why in post:https://stats.stackexchange.com/questions/16493/difference-between-confidence-intervals-and-prediction-intervals
and
they use sqaure root of Variance
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/570269/image.png)
rather than the standard error?? Where is the
?
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/570264/image.png)
댓글 수: 0
답변 (0개)
참고 항목
제품
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!