AIQF Daily digest — 7 new papers

Quant Paper Radar · 3 of 7

1. 1

   Active Portfolio Management in Concentrated Equity Markets

   Brian Ceco, Xiaofei Shi, Ting-Kam Leonard Wong · ARXIV

   we backtested it — Sharpe 0.43 · quant relevance 0.99

   The paper uses forecasts of market concentration and dispersion to allocate dynamically between equal-weighted and capitalization-weighted US equity portfolios. It reports out-of-sample returns after transaction costs. The strategy can be tested on available US equities, but historical S&P 500 membership and the pre-2010 portion of its backtest cannot be replicated.

   original paper

2. 2

   Propose, Don't Judge: An Anytime-Valid Referee for LLM Agents That Mine Investment Factors

   Bo Qu, Mingguang Chen, Licheng Wang · ARXIV

   quant relevance 0.96

   The paper proposes an anytime-valid statistical governance layer for adaptive investment-factor discovery: proposed cross-sectional factors are admitted only when post-submission rank-IC evidence passes online multiple-testing control. A CSI 500 walk-forward study evaluates factor admissions, retirement and portfolio outcomes, including trading costs and Sharpe ratios.

   original paper

3. 3

   Loss Choice or Model Choice? The Role of Forecast Level in Cryptocurrency Volatility Forecasting

   Andrzej Tokajuk, Jarosław A. Chudziak · ARXIV

   quant relevance 0.90

   The paper compares cryptocurrency volatility forecasts across training losses and models, then assesses their implications for one-day Gaussian Value-at-Risk. Its practical finding is that differences between loss functions often reflect persistent calibration of the volatility level rather than better day-to-day volatility dynamics, with implications for risk targeting and VaR-based sizing.

   original paper

4 more papers written up in this period. Open the Quant Paper Radar →