Methodology to test quarterly instruments over the course of a year?

[timmbbo]

Hi there, I'm struggling with how to optimze a strategy, when I run it over 1-2 years.

When I optimize a strategy, I'll run it over Q1, and get a set of best parameters. Then, I'll run over Q2, and get a different set of best parameters. Same for Q3 and Q4. At the end, I'm left with four different sets of best parameters, which I'm not sure how to harmonize together? Or, maybe the fact that I'm coming up with different parameters for different quarters by itself should tell me my strategy is not viable?

I wish there was a way to take a quartely instruments, and perform one test over the course of the entire year to come up with one set of best parameters.

Might you have any thoughts on how to address this challenge? Thanks.

[NinjaTrader_ChelseaB]

Hello timmbbo,

With different data would be different results, which would return a best set of parameters (for that data set).

This thread will remain open for any community members that would like to give opinion on how to interpret these results.

You can run a backtest on a future with rollover periods. Just select the current front contract month, and test for a 1 year range. The merge policy will stitch the rollover periods together.

https://ninjatrader.com/support/forum/forum/ninjatrader-7/strategy-development-aa/96311-backtesting-choosing-contract-month?p=782986#post782986