Daily research update
quant relevance 0.96
The paper develops a covariance-based portfolio method that bridges nested clustered optimization and the unconstrained global minimum-variance portfolio. It argues that partial use of cross-cluster covariance information can improve out-of-sample minimum-variance allocation under estimation error while preserving smaller optimization problems.
quant relevance 0.94
The paper derives dynamic multi-asset position sizing that seeks terminal wealth while enforcing a maximum percentage drawdown constraint. The method can be backtested daily with US equity or ETF prices, but its robust guarantee requires future-return support sets that historical data alone cannot establish.
quant relevance 0.96
The paper develops a spread-aware temporal-hierarchy method for forecasting hourly day-ahead electricity prices and tests its decision value through battery charge and discharge arbitrage. It reports improved forecast accuracy and realized BESS-arbitrage profit gains in German and Spanish power markets.
The rest of the batch
The paper develops a cross-stock conditional diffusion model for next-day equity returns and implied-volatility-surface changes, with applications to option pricing, hedging, scenario generation, and risk management.
The paper derives a robust trading portfolio that maximizes long-run growth under uncertainty in drift and factor dynamics, using an asset covariance structure and an invariant state distribution.
The paper develops an open-source deep-learning model that produces probabilistic buy- and sell-side intraday electricity-price trajectory forecasts from German continuous-market order-book data and neighboring delivery products.
The paper examines whether publicly displayed pre-trade corporate-bond quotes improve liquidity, transaction costs, price dispersion, and price impact, including for OTC executions.
and 10 more · see the full radar
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.