I would like to know if there is a way to get the AIC and BIC of the different orders of the markov chain derived by the command:
fitHigherOrder()
from the markovchain
package.
What I am looking for is something like:
maxOrder <- 5
result <- data.frame()
for (k in 1:maxOrder) {
skip_to_next <- FALSE
mc <- tryCatch(fitMarkovChain(clickstreamList = mydata, order = k), error = function(e) {skip_to_next <<- TRUE})
if(skip_to_next){
next
}else{
result <- rbind(result, c(k, summary(mc)$aic, summary(mc)$bic))
}
}
Use for clickstreams https://bookdown.org/gubiithefish/test/403-Clickstream-application.html
I have tried the same function with fitHigherOrder
and fitHigherOrder()$aic
but they didn't work.
Does it exist a command like assessOrder()
that allows to choose the optimal order when orders are greater than two?