Fellows, I have maternity dataset of about 5000 mothers. I need to analyze the data using hidden markov model (hmm). Can anyone guide me how can I implement this model in R? Do I need to convert my data into TimeSeries so it can fit HMM? If yes how can I convert my excel file data into timeseries data?
It's hard to say what you should do without knowing what your data look like.
Assuming your data has dates and times in it, the R4DS section on working with dates and times should help:
There are several tutorials that should help you get started with HMMs in R:
http://blog.revolutionanalytics.com/2014/03/r-and-hidden-markov-models.html
Hidden Markov Models by Nursultan Svankulov &Eric Friedlander, STOR @ UNC-CH*
Inovance - A Tutorial in R on Using A Hidden Markov Model (HMM)
The vignette for the depmixS4 package might also be helpful https://CRAN.R-project.org/package=depmixS4
6 Likes
Thank you so much for your help. Actually, my dataset contains information of around 5000 pregnant women, it has the name, demographic info, their each pregnancy record such as gender of born child, mode of delivery, diseases etc. I have to run a hidden markov model on this dataset and I dont know what steps should I follow so that my data fits into the model.
PS: As far as I know I need to make a transition matrix which contains probabilities, and I dont know how to compute those probabilities. Any help
Thanks in advance