Back to PLS HelpProblem with adjusted brain scores
Posted on 02/25/09 16:53:28
Number of posts: 60
Dear all,
I don't really know why I have a NAN value in three location in both ulusc-adj and llusc-adj matrix? my orig-usc looks fine but I have a problem with adjusted value.
Does anybody know the reason for that?
please nore: I have performed 100 bootstrap test.
I don't really know why I have a NAN value in three location in both ulusc-adj and llusc-adj matrix? my orig-usc looks fine but I have a problem with adjusted value.
Does anybody know the reason for that?
please nore: I have performed 100 bootstrap test.
Replies:
Untitled Post
I'm Online
Posted on 02/25/09 17:19:31
Number of posts: 291
It means that these 3 location are not stable.
Before we calculate adjusted CI that requires to apply Gaussian Inverse formula, we check how many elements in orig_usc are qualified, which means the bootstrapped value is less than the original value. For elements that are not qualified, we manually set it to NaN. Otherwise, Gaussian Inverse is calculated for each qualified elements. and ulusc_adj/llusc_adj are also calculated.
If you want to know the detail, please check line 1466 to line 1511 of the current version of pls_analysis.m inside Plscmd.zip.
Before we calculate adjusted CI that requires to apply Gaussian Inverse formula, we check how many elements in orig_usc are qualified, which means the bootstrapped value is less than the original value. For elements that are not qualified, we manually set it to NaN. Otherwise, Gaussian Inverse is calculated for each qualified elements. and ulusc_adj/llusc_adj are also calculated.
If you want to know the detail, please check line 1466 to line 1511 of the current version of pls_analysis.m inside Plscmd.zip.
Login to reply to this topic.
