Ever built an angelic trading strategy that performed heavenly in a backtest, only to find it’s a devil in live trading?
Well… there are some very specific “sins” traders make when building trading strategies that destine them (the strategies that is) to a miserable life of soul-sucking underperformance, endless torment and an untimely death.
Nobody wants to see their strategies suffer like that.
So, what can we do to ease the pain and potentially make our strategies better in live trading?
On this BST live show, Rob Carver from systematicmoney.org joins us to discuss “3 ways traders kill trading strategies”. Here are just some of the topics you’ll discover in this episode:
- What can go wrong with strategies in live trading,
- The impact of model complexity on prediction error, (and are simple models really better than more complicated ones – the answer may surprise you!),
- The dangers of overstated backtest performance,
- The impact of historic costs on backtest results,
- The 3 types of overfitting, including 1 that most traders don’t even realise they’re doing, and how to reduce their impact on your trading models,
- Strategy robustness and letting data define complexity,
- The overfitting “levels of sin”,
- Correcting for the “multiple testing problem”,
- Fitting metrics, sample size, data quality, prediction error and much more.
► You can discover more from Rob at http://systematicmoney.org
Don’t forget to subscribe to the BST YouTube channel for all the latest updates and trading content:
Released: Feb 22nd, 2021