I am creating a monte carlo simulation using python that is looking to model divorce rates between different classifications of couples.
To summarize the details:
A marriage can either be an "Arranged Marriage" or a "Love Based Marriage" with a probability assigned to both.
Within each marriage type, there are multiple subgroups of marriages, for simplicity let's say there are two: "Teenage Marriage", "Non-Teenage Marriage", with a probability assigned to both.
They can divorced for two reasons: "Divorce due to infidelity", "Divorce due to something else", with their respective probabilities.
The data is structured like this and goes out for several years:
I need to construct a survival table detailing what month they got divorced and for what reason.
At the moment, I am simply running a loop for each new couple (n=10000) for thousands of trials.
By (1) assigning them randomly to each group based on their respective probabilities, (2) rolling the dice for whether they were divorced and for what reason, then (3) appending the results to a survival table, logging all relevant information.
It bothers me because I feel like this isn't the best way to tackle this problem, but I cannot seem to figure out if there is a more elegant and computationally efficient way to solve this problem. I would sincerely appreciate any insight that anyone could offer.