Back to PLS HelpUntitled Post
Posted on 07/28/08 09:30:49
Number of posts: 29
j crofts
Hi,
I am using the PSL software for the first time and have a brief query.
I am running seed PLS, and although everything (as far as I can tell) is working correctly, I am slightly concerned by a warning from Matlab that it is ignoring the complex part of certain numbers during the comparison stage of the bootstrapping. I guess the bootstapping results cannot be trusted.
Has anyone came across this problem before?
cheers,
Jonathan
I am using the PSL software for the first time and have a brief query.
I am running seed PLS, and although everything (as far as I can tell) is working correctly, I am slightly concerned by a warning from Matlab that it is ignoring the complex part of certain numbers during the comparison stage of the bootstrapping. I guess the bootstapping results cannot be trusted.
Has anyone came across this problem before?
cheers,
Jonathan
Replies:
Seed PLS
Posted on 07/28/08 10:03:00
Number of posts: 394
quote:
Hi,
I am using the PSL software for the first time and have a brief query.
I am running seed PLS, and although everything (as far as I can tell) is working correctly, I am slightly concerned by a warning from Matlab that it is ignoring the complex part of certain numbers during the comparison stage of the bootstrapping. I guess the bootstapping results cannot be trusted.
Has anyone came across this problem before?
cheers,
Jonathan
Hi Jonathan,Hi,
I am using the PSL software for the first time and have a brief query.
I am running seed PLS, and although everything (as far as I can tell) is working correctly, I am slightly concerned by a warning from Matlab that it is ignoring the complex part of certain numbers during the comparison stage of the bootstrapping. I guess the bootstapping results cannot be trusted.
Has anyone came across this problem before?
cheers,
Jonathan
I am happy you are able to get the software running well. I am not entirely sure what you mean by the warning about ignoring complex numbers. Can you:
1. Paste the actual warning
2. Indicate the nature of the data you are analysing.
thanks!
Randy
Untitled Post
Posted on 07/28/08 10:23:44
Number of posts: 29
j crofts
Hi Randy,
Actually, I have just posted that I think I may have solved the problem; maybe you can confirm this.
Basically, I had not removed the seed from the data matrix before running the analysis, which if my
understanding is correct I should have.
cheers,
Jonathan
Actually, I have just posted that I think I may have solved the problem; maybe you can confirm this.
Basically, I had not removed the seed from the data matrix before running the analysis, which if my
understanding is correct I should have.
cheers,
Jonathan
Untitled Post
Posted on 07/28/08 10:32:40
Number of posts: 394
You shouldn't have to zero out the seed voxel since its a small part of the total image volume. Does the error message go away when you zero it out?
Untitled Post
Posted on 07/28/08 10:44:18
Number of posts: 29
j crofts
Yes. I assumed it was probably a numerical error, where the standard error should be zero, but the approximate value is actually very small and negative. Do you have any idea what it might be?
Login to reply to this topic.
