Follow us on LinkedIn
Testing and validating a trading strategy is an important step in trading system development. It’s a commonly-known fact that a well-optimized trading strategy’s performance often deteriorates after it goes live. Thus, developing a robust strategy that performs well out-of-sample is quite a challenge. To this effect, we previously discussed a multiple testing framework for validating a trading strategy. A recent article  looks at the strategy’s robustness from a different perspective; it attempts to answer the question: why a strategy’s performance decays after going live.
To that effect we have proposed two hypothesis. The first one holds that the drop is a pure overfitting effect. In other words, the authors have fine-tuned the (on average) genuine effect to the data used for backtesting in order to make their findings more statistically significant. This finetuning went far beyond what the true effect was and thus only part of the performance is recovered out-of-sample. Second hypothesis posits that once a profitable strategy is disseminated, market participants move in in order to monetize the effect. This in turn weakens or even obliterates the effect going forward.
According to the article, the strategy’s decay is due to 2 effects: overfitting and arbitrage. The authors also pointed out that both effects are equally important,
In order to be able to tell which (if any) of these two hypothesis is more likely, we have constructed a set of explanatory variables for each hypothesis and use univariate regressions to test each variable separately. We find statistically significant coefficients for several variables from both sets. This suggests that both forces are at work and the decay in performance is a joint effect of these forces.
In brief, the article sheds some light on how and why a trading system’s performance deteriorates. However, an important question still remains to be answered: what can we do to mitigate these 2 effects? Let us know what you think.
 Falck, Antoine and Rej, Adam and Thesmar, David, Why and How Systematic Strategies Decay (May 14, 2021). Available at SSRN: https://ssrn.com/abstract=3845928
Have an answer to the questions below? Post it here or in the forum
NTT DATA and AWS cement their credentials as leaders for modernizing mainframe applications PLANO, Texas — NTT DATA, a digital business and IT services leader, announced it is working with Amazon Web Services (AWS) to provide its proven UniKix solution as part of the AWS…