What is walk forward optimisation?

It’s the process of optimising a trading strategy using a limited set of parameters. By testing it on one set of data, and then testing it on a blind set, you can give yourself an indication of whether you're on the right lines.

Where have you heard about walk forward optimisation?

Much has been written about it. One of the biggest problems with developing a trading strategy is that many systems don’t hold up in the future. Walk forward optimisation aims to address this by proving whether your strategy is bad or not.

What you need to know about walk forward optimisation.

The process of walk forward testing isn't difficult. You identify a set of data you want to use to optimise your system. For example, you might want to test data from 2014 to 2015. That's your in-sample data, and then your out-of-sample data might be all of 2016.

You test and analyse your strategy for 2014-2015. You then walk forward to 2016 to review the performance. That's a blind test as you didn’t know how the strategy would perform in 2016 when you tested it in 2014-2015. By doing this, you give it the opportunity to fail.

Find out more about walk forward optimisation.

For background information, read our definitions of out-of-sample and in-sample.