AIQF Daily digest — 16 new papers

Quant Paper Radar · 5 of 16

1. 1

   Long-Horizon Investment with Delayed Information in a Mean-Reverting Market

   Yan Dolinsky · ARXIV · Mean Reversion / Portfolio Optimization

   we backtested it — Sharpe -0.06 · quant relevance 0.91

   The paper derives an optimal investment position for a mean-reverting risky asset when price observations are delayed. Its numerical results illustrate a theoretical model, not measured trading performance on market data.

   original paper

2. 2

   Extending the Oracle Approximating Shrinkage to General Distributions

   Michael J. Lewis · ARXIV · Minimum-variance portfolio construction

   we backtested it — Sharpe 0.53 · quant relevance 0.90

   The paper extends covariance shrinkage to heavy-tailed returns and tests it in rolling global-minimum-variance portfolios of US equities. It reports lower out-of-sample realized volatility than standard OAS, especially in high-VIX periods; it does not establish an improvement in net trading returns.

   original paper

3. 3

   How Much Screening Buys a Second? Speed, Selection, and the Maker's Spread

   Aleksei Adadurov, Eduard Airapetian, Anton Chtepine et al. · ARXIV · Market making and adverse selection

   quant relevance 0.96

   The paper models how faster repricing or screening informed flow changes a liquidity provider’s break-even spread. It calibrates the trade-off using USDC-WETH pool executions and a second-by-second Binance reference price; its reported figures measure spread and adverse selection, not strategy returns.

   original paper

The rest of the batch

• Machine Learning-Based Latency-Aware Smart Order Routing for U.S. Treasury Securities

  Kartik Parmar · OPENALEX · relevance 0.96

  The paper evaluates a latency-aware smart order router for U.S.

• StochLOB: Heavy-Tailed Score Matching for Robust Limit Order Book Trend Prediction

  Arshia Abolghasemi, Seyed Reza Tavakoli, Parsa Naderi et al. · ARXIV · relevance 0.91

  The paper trains a limit-order-book classifier to predict short-horizon mid-price trends and tests its probabilities in an exploratory portfolio application.

and 11 more · see the full radar