the only exception I make for "regime" based models in quant trading
most know how I feel about regime detection (I don't like it), but I'm making an exception for this one.
Python code provided at the end of the article.
💡
🏟️ what do quant funds look for in a backtest?
🕰️ what I do when my ML model is good in one "regime" and bad at others
🚨 thoughts on regime detection/classification etc
🌱 using LightGBM and torch to make it build ML model that's stable across subsamples/regime
🔎 what I learned about model stability after scouring through Kaggle and numerai solutions/forums
🕰️ what I do when my ML model is good in one "regime" and bad at others
🚨 thoughts on regime detection/classification etc
🌱 using LightGBM and torch to make it build ML model that's stable across subsamples/regime
🔎 what I learned about model stability after scouring through Kaggle and numerai solutions/forums
Let's get it.
This post is for paying subscribers only
Already have an account? Sign in.