Smoothed Probabilities in MS model

[Catife]

Dear Tom,

I have a small question about the generated smoothed probabilities from %mssmoothed. After estimating my MS model (converged), using %mssmoothed gives me some probabilities which are not in the interval [0,1]. Some of them are bigger than 1, like 1.0000765262 or even 20. On the other hand, some of them are negative. Could you tell me some possible reasons behind that? I already check the estimated transition probabilities. They look normal and are between 0 and 1. I really appreciate any help from you.

[TomDoan]

No. You would have to post your program and data so I could take a look.