XGB+MLP + two risk models (beta neutral)

Performance Ranking

0.15 98.0th percentile (Sharpe Ratio) 4.11

Better than 98.0% of the 51 other models

Performance Metrics

Primary Rankings

Sharpe Ratio Sharpe Ratio Risk-adjusted return: annualized return divided by annualized volatility. 3.9832

Rank Rank Position on the leaderboard, ordered by Sharpe ratio. #2

Returns

Annualized Return Annualized Return Average yearly return, after scaling the portfolio to the 10% volatility target. This model: Better than 98% of the 51 other models. 39.83%

Best 12M Return Best 12M Return The best compounded return over any 12 consecutive months of the test period. This model: Better than 77% of the 51 other models. 121.27% May 1999 – Apr 2000

Worst 12M Return Worst 12M Return The worst compounded return over any 12 consecutive months of the test period. This model: Better than 98% of the 51 other models. 2.36% Mar 2020 – Feb 2021

Risk

Max Drawdown Max Drawdown Largest peak-to-trough decline during the test period. This model: Better than 69% of the 51 other models. -13.90%

Downside Deviation Downside Deviation Standard deviation of negative monthly returns only. This model: Better than 63% of the 51 other models. 1.675%

Annualized Downside Dev Annualized Downside Deviation Downside deviation in annual terms. This model: Better than 63% of the 51 other models. 5.80%

Risk-Adjusted Performance

Sortino Ratio Sortino Ratio Return per unit of downside risk: like Sharpe, but only negative volatility counts. This model: Better than 84% of the 51 other models. 6.8649

Calmar Ratio Calmar Ratio Annualized return divided by the maximum drawdown. This model: Better than 86% of the 51 other models. 2.8657

Win Rate Win Rate Share of months with a positive return. This model: Better than 96% of the 51 other models. 89.2%

Distribution

Skewness Skewness Asymmetry of returns. Positive skew means more large gains than large losses. 0.102 (Near-symmetric)

Kurtosis Kurtosis Tail thickness. Higher values mean more extreme months, in either direction. 1.014 (Thin tails)

Volatility Targeting

Raw Volatility Raw Volatility Volatility of the portfolio before it was scaled to the 10% target. 2.11%

Scaling Factor Scaling Factor Multiplier applied to the portfolio to reach the 10% volatility target. 4.728

Observations Observations Number of monthly returns used for the evaluation. 408 months

Market Exposure (CAPM)

Beta Beta Sensitivity to market movements. 1.0 moves one-for-one with the market. 0.01

Alpha (ann.) Alpha (ann.) Average yearly return not explained by market exposure (the monthly regression alpha × 12). This model: Better than 96% of the 51 other models. 39.40%

Information Ratio Information Ratio Annualized alpha per unit of residual risk: alpha ÷ std(residuals) × √12. This model: Better than 98% of the 51 other models. 3.9477

R² R² Share of the return variance explained by market movements. 0.6%

Alpha t-stat Alpha t-stat Statistical significance of the alpha estimate. A |t| above about 2 suggests it is not chance. 22.74

Benchmark Comparison

vs. Benchmarks

XGB+MLP + two risk models (beta neutral)

3.9832

Benchmark: Instrumented PCA factor portfolio

1.9481

Benchmark: Kozak-Nagel-Santosh rank-weighted factors portfolio

1.4195

Benchmark: Factor-ML

0.7442

Benchmark: Equal-weight portfolio

0.4912

Benchmark: Minimum Variance

0.4731

Performance Radar

Each axis shows the model's percentile rank within the leaderboard (outer edge = best). Customize the metrics using the dropdowns below.

vs. Leaderboard

Metric	This Model	Average	Top Model	vs. Avg
Sharpe Ratio Sharpe Ratio Risk-adjusted return: annualized return divided by annualized volatility.	3.9832	1.7916	4.1105	+122.3% (better)
Sortino Ratio Sortino Ratio Return per unit of downside risk: like Sharpe, but only negative volatility counts.	6.8649	3.5173	9.7118	+95.2% (better)
Calmar Ratio Calmar Ratio Annualized return divided by the maximum drawdown.	2.8657	1.3100	4.3808	+118.8% (better)
Information Ratio Information Ratio Annualized alpha per unit of residual risk: alpha ÷ std(residuals) × √12.	3.9477	1.7392	4.0976	+127.0% (better)
Win Rate Win Rate Share of months with a positive return.	89.2%	71.6%	90.7%	+24.5% (better)
Annualized Return Annualized Return Average yearly return, after scaling the portfolio to the 10% volatility target.	39.83%	17.92%	41.10%	+122.3% (better)
Max Drawdown Max Drawdown Largest peak-to-trough decline during the test period.	-13.90%	-21.75%	-6.94%	+36.1% (better)

Model Details

Language Python

Competition Year 2024

First Submitted Sep 17, 2026

Resubmitted Sep 27, 2026

Scored Sep 28, 2026

Downloads

Download Code (Python) 66 downloads

Have ideas for making this page more useful? Share your suggestions