Yes thank you, as far as I understood your proposal it would lead to a similar result. Is there a good package in order to do principal component analysis?
Best wishes, Luba ________________________________ Von: Andreas Hary [mailto:andreash...@googlemail.com] Gesendet: Mittwoch, 29. Juli 2009 18:21 An: Stein, Luba (AIM SE) Betreff: Re: [R] Similar package like SEM Are you trying to build forward or yield curves by any chance? Consider using principal components on the log-differences of your yield-curve time series. That sort of gives you something like latent variables. You'll get a bunch of stationary random processes together with factor loadings for your maturities. You can then draw MC-samples from these processes to simulate future yield curves ... if that is indeed what you are after ... Best wishes, Andreas On Wed, Jul 29, 2009 at 1:20 PM, Stein, Luba (AIM SE) <luba.st...@allianz.com<mailto:luba.st...@allianz.com>> wrote: Hello, is there a similar package like SEM. In fact, I am looking for a package where I as input several given vector variables which are part of an autoregressive model. Then I would like to obtan parameters for so called latent variables. Thank you, Luba [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org<mailto:R-help@r-project.org> mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
