AIQF Weekly digest — 109 new papers

The week in numbers

Max drawdown -2.4%, -7.2% at market β · 21 strategies live · as of 2026-10-30

At market β restates the record at the market's beta — the same strategies held at 3.85× the exposure: the return is multiplied by that factor, the drawdown re-measured on the scaled line, and Sharpe is unchanged by it. Arithmetic on the record, not a second track record — no leverage was used and no financing cost is modelled.

Quant Paper Radar · 5 of 109

1. 1

   From pixels to profits: trading arbitrage portfolios based on image representations

   Niklas Paluszkiewicz · OPENALEX · CNN-based Statistical Arbitrage

   we backtested it — Sharpe -0.62 · quant relevance 1.00

   The study uses CNNs to rank stocks by next-month factor-hedged residual returns, using RGB images of stock, hedge, and residual paths. Its strongest reported specification earns 0.82% gross monthly excess return with a 0.97 annualized Sharpe ratio; profitability depends on hedge estimation and trading costs.

   original paper

2. 2

   Assessing the effectiveness of Greenblatt's Magic Formula across international stock markets : literature review

   Podgórski, Krzysztof · ECONBIZ · Quality and Value

   we backtested it — Sharpe 0.38 · quant relevance 0.99

   This literature review evaluates an annually rebalanced, equal-weight equity strategy combining high return on tangible capital with high EBIT/enterprise value. It synthesizes published backtests rather than presenting a new original backtest, highlighting transaction costs, universe selection and whether earnings yield alone performs better. The US implementation is buildable over available history, but the international results and earlier sample periods cannot be replicated with platform coverage.

   original paper

3. 3

   Sparse Autoencoders as Implicit Diversification Regularisers for Cross-Sectional Cryptocurrency Returns

   Weijia Han, Lisha Qu · ARXIV · Sparse Factor Diversification / Cross-Sectional Mean Reversion

   we backtested it — Sharpe 0.26 · quant relevance 0.98

   Sparse-autoencoder reconstructions produce contrarian portfolios with lower measured volatility than matched-dimension PCA, but do not establish superior risk-adjusted returns or stable factor identities. The return-based method is implementable on the secondary US-equity market, although a practical test must replace the study's explicitly non-executable same-close entries with subsequent-price execution.

   original paper

4. 4

   Optimal portfolio choice under green transition timing risk

   Emma Kroell, Mogens Steffensen · ARXIV · Robust Portfolio Optimization

   we backtested it — Sharpe 0.72 · quant relevance 0.97

   The study derives robust portfolio allocations when an uncertain, one-time green transition raises green-stock prices and lowers brown-stock prices. Its allocation rule can be tested on US equities using scenario-assumed transition probabilities and jump sizes, but historical returns cannot validate the assumed transition mechanism. The supplied material presents theoretical results and numerical illustrations, not measured trading performance.

   original paper

5. 5

   A Drift-Aware Continuous Hybrid Learning Framework for Adaptive and Robust Stock Trading

   Annalakshmi N, S Thangapriya · OPENALEX · Adaptive Machine Learning / Reinforcement Learning

   quant relevance 1.00

   The framework combines adaptively weighted SVM and XGBoost forecasts with a DDPG trading policy and drift-triggered retraining. Its US-equity, daily-frequency implementation is buildable, but NIFTY 50 coverage is unavailable and historical S&P 500 membership requires a disclosed capitalization-screen approximation. The abstract reports improved risk-adjusted trading performance, although numerical results are not visible in the supplied excerpt.

   original paper

109 new papers written up in this period. Open the Quant Paper Radar →