Skip to content

Conversation

@t2o2
Copy link
Contributor

@t2o2 t2o2 commented May 8, 2022

Fixes #715
Closes #473

  • Remove flooring on Sharpe Ratio
  • Adding Kelly Criterion

@t2o2
Copy link
Contributor Author

t2o2 commented May 13, 2022

Any other comments?

@cscbai
Copy link

cscbai commented Jun 9, 2022

Any update on this review?

@kernc
Copy link
Owner

kernc commented Jun 9, 2022

I've actually been meaning to but just didn't come round to trying it out and seeing what happens when positions are indeed sized as Kelly Criterion recommends. Say I run a backtest and get Kelly Criterion=0.5. Say I now repeat the same strategy backtest with trades size=0.5. What does the Kelly Criterion say at the end of this backtest? And if the value is different from before, of what real use is it?

@t2o2
Copy link
Contributor Author

t2o2 commented Jun 9, 2022

The traditional meaning of Kelly Criterion is indeed the bet size optimisation parameter. However, it also shows the confidence level on a particular strategy, similar to the Sharpe ratio, the higher is generally better.

In a portfolio sense, it tells the user, how much to invest in this particular strategy as part of user's overall portfolio.

@t2o2
Copy link
Contributor Author

t2o2 commented Oct 12, 2022

Is it possible to merge in?

@kernc kernc force-pushed the master branch 9 times, most recently from 60eff81 to 109c352 Compare November 28, 2022 22:33
@kernc kernc changed the title Remove flooring on ratios and adding Kelly Criterion Remove flooring on ratios & add Kelly Criterion Dec 5, 2022
@kernc
Copy link
Owner

kernc commented Dec 5, 2022

Sorry to have taken so long. CI was a mess. Thanks for the PR! 🍰

@kernc kernc merged commit d47185f into kernc:master Dec 5, 2022
Goblincomet pushed a commit to Goblincomet/forex-trading-backtest that referenced this pull request Jul 5, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Negative Sharpe Ratio!

3 participants