I have a dataset consisting of patients and their equidistant visits and I have labelled the presence of a specific kind of mole in their left and/or right hand with {0,1} values (0 = not present and 1 = present). The dataset looks like this:
R L
Patient 1 0 1
Patient 1 1 1
Patient 1 0 1
Patient 1 0 1
Patient 1 0 1
Patient 2 1 1
Patient 2 0 1
Patient 2 0 1
Patient 2 1 1
Patient 3 0 0
Patient 3 1 1
Patient 3 0 0
Patient 3 0 1
Patient 3 1 1
Patient 3 0 1
So, that means that patient 1 had 5 visits with the presence of mole identified in his hand by Yes or No signified by 1 or 0.
I am interested in finding the probability of 1) a patient and 2) a hand (left or right) developing a mole in a) general, b) in a specific visit and c) in a future visit.
The first though that came to me is to model this is with Markov Chains since the visits are equidistant and I can introduce a higher order in order to make the Chain having a kind of memory. Now, I came into the package LMest of R and I was wondering if you think it's suitable for my modelling purposes.
Any ideas?
