walk forward analysis

Posted on Apr 19, 2020 by Kris Longmore
5 comments.
0 Views

Optimisation tools have a knack for seducing systematic traders. And what’s not to love? Find me the unique set of parameters that delivered the greatest return in my ten-year backtest. And do it in under five seconds. That’s certainly attractive. But do you want to hear something controversial? When it comes to the parameters of a systematic trading strategy, in the majority of cases optimisation is your enemy. A siren, luring you to shipwreck. On the other hand, there are a few applications of optimisation that are helpful and sensible in a trading context. Usually however, they’re not obvious. So we decided to write a series of blog posts to give you the no-holds-barred low-down on optimisation for systematic trading. We’ll start with a lightning tour of various optimisation techniques (this article) before moving on to: The implicit assumptions you make when you optimise strategy parameters Best practices related to optimisation of strategy parameters Optimisation use cases in the context of systematic trading Optimisation tools and techniques Let’s start with a whirlwind tour of a selection of optimisation tools. We’ll...

Posted on Nov 24, 2015 by Kris Longmore
5 comments.
0 Views

Important preface: This post is in no way intended to showcase a particular trading strategy. It is purely to share and demonstrate the use of the framework I've put together to speed the research and development process for a particular type of trading strategy. Comments and critiques regarding the framework and the methodology used are most welcome. Backtest results presented are for illustrating the methodology and describing the outputs only. That done, on to the interesting stuff My last two posts (Part 1 here and Part 2 here) explored applying the k-means clustering algorithm for unsupervised discovery of candlestick patterns. The results were interesting enough (to me at least) to justify further research in this domain, but nothing presented thus far would be of much use in a standalone trading system. There are many possible directions in which this research could go. Some ideas that could be worth pursuing include: Providing the clustering algorithm with other data, such as trend or volatility information; Extending the search to include two- and three-day patterns; Varying the number of clusters; Searching across markets and asset...