The GT-Score for Reducing Overfitting in Trading Strategy Optimization
Summary
This document presents the GT-Score, a composite objective for optimizing data-driven trading strategies. It combines measures of performance, statistical significance, consistency, and downside risk, aiming to discourage choices that fit historical data too closely. The motivation is that repeated testing can produce spurious patterns and that common statistical inference may be unreliable when returns are not normally distributed.
The empirical study uses historical data for 50 S&P 500 companies from 2010 to 2024. It evaluates three strategies with nine sequential walk-forward splits and a Monte Carlo study using 15 random seeds. The authors report a higher validation-to-training return ratio under GT-Score than under baseline objectives, while paired tests find detectable differences from Sortino and Simple objectives with small effect sizes. These results support further study of the objective, but the evidence is confined to the stated stocks, strategies, and evaluation design. A better generalization ratio alone does not guarantee profitable deployment or resolve every source of model and execution risk.
Key ideas
- The GT-Score combines performance, significance, consistency, and downside-risk considerations.
- Its design targets data snooping and weak inference under non-normal returns.
- The evaluation uses walk-forward validation and a Monte Carlo study across three strategies.
- Reported differences include small effect sizes, and the evidence covers a limited historical sample.
Tags
Full text
# The GT-Score: A Robust Objective Function for Reducing Overfitting in Data-Driven Trading Strategies # The GT-Score: A Robust Objective Function for Reducing Overfitting in Data-Driven Trading Strategies Overfitting remains a critical challenge in data-driven financial modeling, where machine learning (ML) systems learn spurious patterns in historical prices and fail out of sample and in deployment. This paper introduces the GT-Score, a composite objective function that integrates performance, statistical significance, consistency, and downside risk to guide optimization toward more robust trading strategies. This approach directly addresses critical pitfalls in quantitative strategy development, specifically data snooping during optimization and the unreliability of statistical inference under non-normal return distributions. Using historical stock data for 50 S&P 500 companies spanning 2010-2024, we conduct an empirical evaluation that includes walk-forward validation with nine sequential time splits and a Monte Carlo study with 15 random seeds across three trading strategies. In walk-forward validation, GT-Score improves the generalization ratio (validation return divided by training return) by 98% relative to baseline objective functions. Paired statistical tests on Monte Carlo out-of-sample returns indicate statistically detectable differences between objective functions (p < 0.01 for comparisons with Sortino and Simple), with small effect sizes. These results suggest that embedding an anti-overfitting structure into the objective can improve the reliability of backtests in quantitative research. Reproducible code and processed result files are provided as supplementary materials.
Shown in full with attribution under the source's licence. Licence: abstract CC0
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.