Bayesian Testing of Trading Ideas

Conviction should follow confirmation. The only confirmation I see discussed around trading on Twitter are additional untested signals. My goal is to change that, at least a little bit.

My goal is to promote scientific skepticism in the trading world. I apply Bayesian hypothesis testing to quantify the uncertainty around return distributions following so-called signals. If the signals have edge, the return distributions will show edge.

I believe that backtesting should be the last step in evaluating a strategy. Results could easily be attributable to luck. Instead, I assume there is no effect and allow the data to change my mind via a Bayesian updating of my priors.

It’s the way we naturally think, and it’s the way we test medicine. Why don’t we put our money where our mind is?

Sign up now so you don’t miss the first issue.

Please follow me on Twitter!