Hi,
although there is already an entry concerning the bootstrap error: "A singular matrix occured...." I have made an interesting observation which I want to share with the smartPLS developers. It seems that XLSTAT can run easily bootstrapping (dont worry, I use mainly smartPLS) even with a block of dummy variables. SmartPLS on the óther hand aborts already with the first drawn sample. So it seems that there is a high probability that smartPLS draws identical samples or?