I'm hoping some experienced futures traders will comment on this post.
I would like to develop some systematic strategies to trade futures after many years developing systems for stocks, and have a few questions.
What is the recommended way to back-adjust the historical futures data for backtesting? ratio-adjusted, or difference-adjusted, or something else?
What is the recommended way to measure P&L? I'm accustomed to calculating returns as a percentage of account equity, but I wonder if futures traders should measure P&L in ticks.
Should backtests use fixed size positions or compounding? I have always compounded returns in my equity backtests.
Many thanks in advance.
I do not believe in right , wrong or recommended way when you talk about trade. I believe in comparing results and use what it is best.
Let's say your system fits well with ratio-adjust, why you will use other type of adjust ?
Amibroker is so flexible that it's worth testing the various combinations
@awilson Thank you for the reply. I agree AmiBroker has the flexibility to test various combinations, however it's unclear how to decide which combination is best. There is a risk of introducing distortions in the backtest results and I want to avoid fooling myself by choosing an approach that produces unrealistic results. Distorted results are bad, but being unaware of the distortions is even worse. Since I'm inexperienced with futures, there are probably some new backtesting pitfalls I've never encountered with stocks and ETFs.
If that happen , you will find out during paper work phase (demo account) and this is part of your learning process.
The best advice I can give you regarding futures is :
- Do not be in a hurry to start operating.
- and Test, test, test,
@pmxgs Thank you for the helpful reply, and as @awilson mentioned, testing is essential.
For the sake of others who may find this thread later, I've done some searching online and found a related thread on another website. Hope it's okay to post the link here.