[R] Markov Decision Process

[Christian Schulz]

Maybe  msm (Multi-state modelling)  is a starting point. You'll find the 
manual in the package/doc folder after installation.

HTH
Christian


> Dear All,
> I am struggling with the conceptual aspects of a problem.
> I am sure that someone on this list must be familiar with this.
> Let's say that you have some cancer data for your patients.
> In particular, every patient may undergo up to [i.e. the cycles may 
> stop earlier for various reasons] 6 cycles of therapy (hormonal or 
> chemotherapy) whose durations and starting times are known. There are 
> plenty of other data available, but let us keep it simple for now.
> At the end of the therapy cycles, you know if the patient is dead or 
> alive (in reality, the final states are more as the patient may be 
> dead with/without cancer or alive with/without cancer, but again, 
> let's keep it simple for now).
> Of course, you want to develop a policy which maximizes the 
> probability of the patient to be alive at the end of the cycles of 
> therapies.
> Does anybody know how to tackle this in a Markov decision approach?
> There are so many R packages dealing with Markov chains that it is 
> almost confusing for a beginner.
> Any suggestion is welcome.
> Many thanks
>
> Lorenzo
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide 
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>