1

I work in fin-tech and would like to build some sort of simulation program to assess how different inputs will impact net revenue. For example, if we create new policies based on ML scores, how would those have impacted our loss and revenue metrics?

While we can and do run online experiments, it would be desirable to simulate these impacts ahead of time.

Aside from something like reinforcement learning, I was thinking that Monte Carlo simulations might be the best approach. Anyone do something similar before or have suggestions?

Kevin
  • 11
  • 1
  • can you do a backtest on historical data? – oW_ Jul 22 '20 at 00:12
  • Yes, thats possible but not the issue I have. I'm looking for guidance on how one would simulate it. – Kevin Jul 22 '20 at 23:16
  • Maybe you can add some color on what you would like to accomplish that you can't get from a backtest. "For example, if we create new policies based on ML scores, how would those have impacted our loss and revenue metrics?" - this sounds like something a backtest could answer. I think this could be helpful for others to answer your question. – oW_ Jul 22 '20 at 23:32

0 Answers0