System Parameters Randomization, better than “System Parameters Permutation”

Hello,

There is an interview with Dave Walton, creator of SPP (System Parameters Permutation) where he talks about a better way of approaching this .

The idea is to take the SPP and put a Monte Carlo function on top of it. Monte Carlo selection engine overlaid on top of SPP.

Here is the paper by Dave Walton:
http://bettersystemtrader.com/system-parameter-permutation-a-better-alternative

“In order to address all of these, I’d like to introduce a close cousin of SPP which I named System Parameter Randomization (SPR). The fundamental mechanism behind both SPP and SPR is the same, yet the implementation is very different. SPR can be thought of as a random sample of a continuous SPP distribution. The process is explained in the steps below:

1. Parameter scan ranges for the system concept are determined by the system developer.
2. A Monte Carlo process is used to pick individual parameter values within in the scan ranges.
3. A fixed number of iterations is performed using a realistic portfolio-based historical simulation over the selected time period.
4. The simulated results for each system variant are combined to create a sampling distribution for each performance metric of interest (e.g. CAR, max drawdown, Sharpe ratio, etc.). Each point on a distribution is the result of a historical simulation run from a single system variant.”

Full interview:
http://bettersystemtrader.com/051-dave-walton/

Hello,

that’s a good point, thanks for sharing! I have notified Mark about this to check …

https://roadmap.strategyquant.com/tasks/sq4_8737

Hi all,

Has this already been implemented in SQX?
Thank you