AQAI QuantAI research lab for systematic strategies

Daily research update

Daily digest — 13 new papers

Sent 3 September 2026 · 13 new papers

Quant Paper Radar · 4 of 13

  1. 1

    Insights on Time-consistent Deep Hedging under Elicitable Dynamic Risk Measures

    quant relevance 0.97

    The paper develops a reinforcement-learning method for daily dynamic hedging of a high-dimensional basket option under time-consistent conditional CVaR objectives. Its numerical results are based on a simulated DCC-GARCH environment rather than a measured real-market trading backtest, but the hedging method can be tested on historical US-equity baskets with synthetic option liabilities.

    original paper

  2. 2

    Eliciting ESG Preferences for Reinforcement Learning-Based Portfolio Optimization

    quant relevance 0.96

    The paper proposes daily ESG-aware portfolio optimization using a preference-conditioned PPO policy, with Sharpe ratio and three agency-specific ESG scores as objectives. A Gaussian-process preference-elicitation layer infers portfolio-manager objective weights from pairwise portfolio choices, evaluated using LLM personas under regional prompts. It is directly relevant to systematic portfolio construction, but the required multi-provider ESG data are not available.

    original paper

  3. 3

    Price manipulation in nonlinear transient impact models: rigidity before memory and complete positivity after memory

    quant relevance 0.96

    This is a theoretical market-microstructure and optimal-execution paper characterizing when nonlinear transient price-impact models permit profitable round-trip manipulation. Its central practical implication is model selection for execution-cost estimation and algorithmic trade scheduling: concave impact should be applied after resilient memory rather than directly to trading rate unless sufficient friction is included.

    original paper

The rest of the batch

and 9 more · see the full radar

This issue was emailed to our subscribers. Subscribe to get the next one — daily or weekly, your choice.

Backtest and model results are research artifacts, not live trading results and not a guarantee of future performance. Informational and educational purposes only. Not individualised investment advice.