Thanks, yes a good and simple starting point, how to integrate Bayes should my own challenge.
christian ----- Original Message ----- From: "Spencer Graves" <[EMAIL PROTECTED]> To: "Christian Schulz" <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Thursday, September 18, 2003 4:25 PM Subject: Re: [R] 2 time dependend states and probability transition matrix > I don't know if I understand what you want, but I wonder if the > following might help: > > > Ptrans <- data.frame(Jan=LETTERS[rep(1:2, 5)], > + June=LETTERS[rep(1:2, each=5)]) > > (table(Ptrans$Jan, Ptrans$June)/dim(Ptrans)[1]) > > A B > A 0.3 0.2 > B 0.2 0.3 > > hope this helps. spencer graves > > Christian Schulz wrote: > > >Hi, > > > >have got anybody experience or a starting > >point for me, how i can program a function > >which calculate me a probability transition matrix with a data.frame and two states. > > > >I have some independend variable and a class variable > >for two states - in example one from Jan2003 and the same variables from June2003. > > > >Now i would like to calculate the probability that PERSON X change > >from classA to classB between this time-period dependend from > >the independend variables. > > > >Until now i think about an "easy" function what use the Bayes-Theorem, but it's too bad that i'm > >not a mathematican. > > > >Thanks for any starting point, idea, link or paper! > > > >P.S. > >I now the msm package, but think it's complex for my intention!? > > > > > >Regards,Christian > > > > > > > > [[alternative HTML version deleted]] > > > >______________________________________________ > >[EMAIL PROTECTED] mailing list > >https://www.stat.math.ethz.ch/mailman/listinfo/r-help > > > > > > ______________________________________________ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help ______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
