AIQF Daily digest — 18 new papers

Quant Paper Radar · 8 of 18

1. 1

   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

2. 2

   Reassessing Reinforcement Learning for Optimal Trade Execution

   Matteo Tosi, Enrico Barbierato, Alice Gatti · OPENALEX · Optimal Execution / Residual Reinforcement Learning

   quant relevance 1.00

   The study evaluates PPO execution policies against TWAP on paired historical intraday paths and finds that bounded residual adjustments stabilize liquidation schedules without statistically reliable execution improvements. Its sparse-state, parameterized-cost simulator is implementable with minute bars, but that would test simulated execution rather than actual quote-level execution quality.

   original paper

3. 3

   Weekend Crypto Volatility and Monday Market Risk

   Pedram Ebadollahyvahed, Nneka Umeorah · ARXIV · Calendar-Based Volatility Forecasting and Risk Targeting

   we backtested it — Sharpe 0.35 · quant relevance 0.94

   Weekend cryptocurrency volatility improves forecasts of Monday variance across equities, gold and a Treasury ETF, primarily after 2023. This supports a testable Monday position-sizing overlay, but the reported forecast improvements are not trading returns, the mechanism remains unresolved, and our shorter post-2024 sample limits validation.

   original paper

The rest of the batch

• Simple versus optimized ETF portfolios: Does complexity pay?

  Michal Mešťan · OPENALEX · relevance 1.00

  The study compares annually rebalanced fixed stock–bond allocations, equal weighting, capped inverse-volatility weighting, and constrained Markowitz portfolios.

• FinLLM-KG: Knowledge Graph-Enhanced LLMs for Financial Sentiment Analysis in Algorithmic Trading

  Giorgos Iacovides, Oliver Hannibal, Wuyang Zhou et al. · ARXIV · relevance 0.99

  FinLLM-KG augments financial-news sentiment scoring with temporally eligible company facts extracted from SEC 10-K filings, then ranks stocks into daily long–short portfolios.

• THE ROLE OF CRYPTO ASSET DIVERSIFICATION IN REDUCING RETAIL INVESTOR PORTFOLIO RISK

  Adang Sutisna, Nur Fitriyani, Arasy Alimudin et al. · OPENALEX · relevance 0.98

  The study evaluates small, monthly rebalanced crypto allocations in Indonesian stock, government-bond, and gold portfolios, distinguishing absolute risk from risk-adjusted returns.

• The Term Structure of Implied Fee Rates in Liquidity Pools

  Diogo Duarte, Lucas X. Ferreira, Emmanuel Gobet et al. · ARXIV · relevance 0.96

  The study values Uniswap ETH–USDC liquidity-provider payoffs using option-implied risk-neutral dynamics to estimate fee compensation across holding horizons and liquidity ranges.

• Does Ecosystem Activity Validate TVL?

  Matthew Brigida · ARXIV · relevance 0.96

  The study tests weekly crypto portfolios sorted jointly on TVL-to-market-cap and ecosystem activity, finding no incremental return signal robust to multiple-testing adjustment.

and 10 more · see the full radar