AIQF Daily digest — 5 new papers

Quant Paper Radar · 5 new

1. 1

   Optimal Trading Filters and the Value of Predictive Signals

   Matteo Smerlak · ARXIV · Predictive-signal portfolio optimization

   we backtested it — Sharpe 0.07 · quant relevance 0.98

   The paper derives causal position-sizing filters for return forecasts under inventory risk and transient trading costs. Its results are theoretical, not a reported backtest. A daily US-equity implementation is feasible, but its assumed price-impact kernel cannot be validated from bar data alone.

   original paper

2. 2

   Rachev-Ratio Portfolio Optimization with TimeGAN Scenarios: Replacement versus Supplementation

   Yifan He · ARXIV · Tail-Risk Portfolio Optimization

   we backtested it — Sharpe 0.48 · quant relevance 0.98

   The study directly tests long-only ETF allocation using a tail-sensitive Rachev-ratio objective with historical, TimeGAN-generated, and hybrid scenarios. Generated scenarios distort tails and dependence, and paired inference does not establish a reliable investment-performance advantage. The method is buildable, but available prices require a later evaluation start after accumulating the 756-day training window; the original 2007–2025 dataset cannot be reproduced in full.

   original paper

3. 3

   Expected utility maximization and the optimal portfolio of risky assets

   Whelan, Karl · ECONBIZ · Portfolio Optimization

   we backtested it — Sharpe 0.60 · quant relevance 0.94

   The paper derives a utility-based allocation to risky assets proportional to the tangency portfolio, with a specific adjustment to overall risky exposure. It reports numerical utility-efficiency comparisons using empirical return data, rather than a realized trading backtest.

   original paper

The rest of the batch

• Dynamic Programming for Value-at-risk Markov Decision Processes in Continuous Spaces

  Xian Chen, Erick Delage, Weikai Wang · ARXIV · relevance 0.95

  The paper establishes dynamic programming equations for continuous-space VaR objectives and applies them to adaptive optimal execution.

• Estimation of high-dimensional volatility matrices with dynamic conditional correlation-embedded mixed factor structures

  Dai, Runyu, Matsuda, Yasumasa · ECONBIZ · relevance 0.94

  DCC-ePOET combines observable factors, sparse weak latent factors, and dynamic residual correlations to estimate high-dimensional conditional covariance matrices.