Hi, all. There are a few alogorithms that can do a PCA transform on gappy data. Just to name a few works : Connolly & Szalay, Everson & Sirovich, etc...
I was wondering if anybody ever implemented one of these or similar algorithms in S-Plus or R... I tried to do this myself and although it does not look extremely complicated I spent quite a lot of time debugging the code... whithout success. If anybody ever ran into some working code on the subj, plz let me know. BTW, it occured to me that if S-Plus could do regression with multiple dependent variables (which is what Connolly & Szalay method is basically based on) it would be trivial to implement the gappy PCA algorithm... Unfortunately I could not find any such algorithm either... . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
