[ssml] About missing data in principle component analysis

[xueping quan]

Dear All
    I have about 200 samples with 90 variates (then a 200*90 matrix) and want to do principle component analysis using MATLAB. It should be very easy but there are some missing data in this matrix and MATLAB shows error message that it can not converge these data. SAS can deal this problem but it can not handle so many variates. Does anyone has some idea about how to handle this problem with MATLAB? Thanks a lot!
 
Cheers
 
xueping

		
---------------------------------
Yahoo! Messenger NEW - crystal clear PC to PCcalling worldwide with voicemail
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://bioinformatics.org/pipermail/ssml-general/attachments/20050606/652ab269/attachment.htm