I have 10 different states of interest.
People have a set number of chronic conditions. In the simplest model, I am interested in how people transition from one disease count to the next.
People can have x = {0,1,...,9} number of conditions. People can move 0-1-2-3-4-5, but also 0-1-4-5, or 0-4-2-3-5 (because their disease count can decrease in a particular version of the example that allows for recovery).
A sequential logit, as is, does not seem appropriate because people have to move through specified paths (e.g. 0:1 2 3; 1:2 3; 2:3).
If I want to model the transition probabilities between each state, is there some joint procedure available? It would seem to me that there are 1024 (2^10) possible transitions to estimate in this case?
People have a set number of chronic conditions. In the simplest model, I am interested in how people transition from one disease count to the next.
People can have x = {0,1,...,9} number of conditions. People can move 0-1-2-3-4-5, but also 0-1-4-5, or 0-4-2-3-5 (because their disease count can decrease in a particular version of the example that allows for recovery).
A sequential logit, as is, does not seem appropriate because people have to move through specified paths (e.g. 0:1 2 3; 1:2 3; 2:3).
If I want to model the transition probabilities between each state, is there some joint procedure available? It would seem to me that there are 1024 (2^10) possible transitions to estimate in this case?
Comment