Journal of Investment Strategies

Risk.net

Statistical testing of DeMark technical indicators on commodity futures

Marco Lissandrin, Donnacha Daly and Didier Sornette

  • The focus is to understand what to expect from the price development, just after a completed signal.
  • This is done by sliding number of holding days in combination with permutation tests.
  • In most cases there is a limited range of holding days for which an indicator has predictive power.
  • Significantly exceptional price moves might come with a delay and they might last only for a few days.

In this paper, we examine the performance of three DeMark indicators (Sequential, Combo and Setup trend), which constitute specific implementations of technical analysis often used by practitioners, over twenty-one commodity futures markets and ten years of daily data. Our work addresses price behavior following new entry signals by studying whether, for short holding periods, the entry signals generated by these indicators can time the market moves and suggest the right side of the market (long or short). For example, we want to know how long we should hold or delay a trade before the price is expected to move significantly. The signals are sparse, as they mostly suggest entering the market between one and five times per year. To adjust for the limited number of total days in which the trade signals are in-the-market, we generate the distributions of multiple performance metrics (mean return, profit factor and risk–return ratio) over different trade holding horizons, and compare them with their randomized versions, which have the same number of entry signals and the same number of holding days. The rolling strategy, which creates continuous futures data from separate contracts, plays a role in evaluating the statistical performance of these indicators. Overall, this paper gives more clarity to the predictive performance of these indicators as well as practical guidance on how to use them in a trading environment as generators of market entry ideas.

To continue reading...

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an indvidual account here: