It seems that indeed providing other starting values initiates iterations
to take place. However, more worrisome is that the does not seem to
converge, even when upping the number of iterations. Below I run a 2 state
model on the same as well for comparison (I have added set.seed statements
to make
I think it's your starting values for the initial state probability
distribution,
i.e. c(1,1,1)/3 that cause the problem. They seem to drop you into some
sort of local maximum/stationary point, a long way from the global maximum.
Try, e.g. c(4,2,1)/7; this gives me:
hmmFit$hmm$emissionProbs
Hi,
I am having difficulties estimating the parameters of a HMM using the HMM
package. I have simulated a sequence of observations from a known HMM. When
I estimate the parameters of a HMM using these simulated observations the
parameters are not at all close to the known ones. I realise the estim
3 matches
Mail list logo
