Dear Quantopian community,

I only discovered Quantopian 2 to 3 weeks ago and am since then experiencing the greatest learning lessons around finance topics I've seen so far. You've done a supergreat job at explaining finance topics theoretically and then even provide the code to experiment around for yourself. Wow. Thank you!

But currently I reach the boundaries of your learning material and I have questions I could not find answers so far, neither on Quantopian nor elsewhere. One of these questions is around quantifying risk and optimizing your trading strategy around it.

My stomach feeling tells me that there should be a way to formalize risk as a probability of ruin within a time horizon (let's say 21 days), given your current level of wealth and (assuming normally distributed returns of your portfolio) mu and sigma, rather than VaR or CVaR. I tried to search for it and found the following paper that talks about the concept in an insurance setting: https://www.researchgate.net/publication/311525074_Minimizing_expected_time_to_reach_a_given_capital_level_before_ruin
It mentions a paper from 1985 that sounds more appropriate for what I am looking for: Continuous-Time Red and Black: How to Control a Diffusion to a Goal:https://pubsonline.informs.org/doi/abs/10.1287/moor.10.4.599. I am confused that the condition only depends on mu and sigma, but not the current level of wealth?
And in Feller "Introduction to Probability Theory and Its Applications" volume 1 chapter XIV.6 he talks about Fürth's formula for first passage.

Does anyone on this forum know of an accessible explanation of an approach that quantifies risk rather in terms of probability of ruin than in terms of VaR/CVaR?

Thanks a lot!
Christian