필터 지우기
필터 지우기

Confidence levels for glmfit - glmval or fitglm using the Likelihood Ratio method

조회 수: 5 (최근 30일)
Craig
Craig 2024년 4월 27일
댓글: Craig 2024년 5월 6일
After several days and many hours of researching and googling, I am unable to determine how I could find confidence levels using the Likelihood Ratio for either glmfit - glmval or fitglm. I know it can be done in R, but I don’t see that capability in Matlab. Does it exist? Neither glmval nor coefCI appear to use or have options for using the Likelihood Ratio method. And paramci does not appear to work with a GeneralizedLinearModel object.
Thanks for any help you can provide.

답변 (1개)

Akshat
Akshat 2024년 5월 3일
Hi Craig,
Upon looking up the documentation for the relevant functions which can be used to serve your functionality, I unfortunately couldn't find anything that finds the confidence levels using the Likelihood ratio.
On the other hand, I did see that "coefCI" uses the Wald Method to determine the confidence intervals, which as far as I could find, is an approximation of the Likelihood Ratio test.
Now as a workaround as you said, there is a functionality in R for the same, I can suggest you the following:
This is a filexchange developed to create a link between MATLAB and R, using which you can get the variables from R, and also run scripts there. This can help you keep the entire pipeline in MATLAB, with just the confidence intervals being calculated in R.
Hope this helps!
  댓글 수: 1
Craig
Craig 2024년 5월 6일
Hello Akshat, Thank you for looking at this. I had found your reference, and it was helpful, but as you say, the Wald Method is just an approximation to the Likelihood Ratio method.
I was not aware of the MATLAB R-link file exchange package. That might be a work around, it would require me to reverse engineer the R-code I have, but it is a possibility.
I'm going to leave this open for now to see if anyone can provide additional assistance.
Again, thanks for your helpful reply.

댓글을 달려면 로그인하십시오.

카테고리

Help CenterFile Exchange에서 Analysis of Variance and Covariance에 대해 자세히 알아보기

제품


릴리스

R2023a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by