Weekly research digest
| AI Quant portfolio | AI Quant | S&P 500 |
|---|---|---|
| Week to 8 Sep | +0.3% | +0.5% |
| Since inception | +12.1% | +14.5% |
| Since inception, at market β | +44.8% | +14.5% |
| Sharpe | 3.14 | 2.01 |
| Beta | 0.27 | 1.00 |
At market β restates the record at the market's beta — the same strategies held at 3.70× the exposure: the return is multiplied by that factor, the drawdown re-measured on the scaled line, and Sharpe is unchanged by it. Arithmetic on the record, not a second track record — no leverage was used and no financing cost is modelled.
To explain the pipeline, we'll follow Hybrid Overnight Momentum with Intraday Extension Fade, a strategy from our research catalogue that combines overnight return persistence with an intraday reversal signal.
A monthly risk overlay measures persistent rotation in the subdominant correlation eigenspace of a point-in-time large-cap stock universe. When REC improves completed out-of-sample SPY variance forecasts, positive REC readings reduce SPY exposure in favor of cash.
The strategy holds a long NEPSE Index exposure scaled to a 12% annualized volatility target using one-day Student-t GARCH or EGARCH forecasts. It shortens the estimation window and halves exposure when point-in-time stability diagnostics flag an unstable regime.
A point-in-time correlation-network indicator shifts a large-cap equity portfolio into a 75% MBB defensive allocation when systemic co-movement reaches an expanding-history extreme. Hysteresis and asymmetric signal memory reduce rapid state reversals.
we backtested it — Sharpe 0.75 · quant relevance 1.00
The paper proposes and backtests a daily, market-neutral US equity-selection strategy using technical, fundamental, macro and text-sentiment data. The core framework can be implemented with available data, but its search-attention inputs and long pre-2020 alternative-data history cannot be replicated.
we backtested it — Sharpe 0.78 · quant relevance 0.98
The paper compares deep-reinforcement-learning portfolio allocation with classical construction methods across rolling market-efficiency regimes. Its tradable idea uses a fuzzy autoregression-based inefficiency measure to select or condition the portfolio optimizer and risk budget.
we backtested it — Sharpe 0.24 · quant relevance 0.98
The paper develops parametric EVaR portfolio optimization under multivariate tempered-stable Lévy return models and tests minimum-risk and entropic reward-risk allocations in US sector ETFs. In rolling out-of-sample results, several entropic portfolios achieve higher realized Sharpe ratios than matched CVaR portfolios and standard allocation benchmarks.
we backtested it — Sharpe 0.61 · quant relevance 0.98
The paper derives market-timing and short-horizon portfolio-ranking rules from state-price densities recovered from option-implied-volatility smiles. Its empirical comparisons show that the timing rules generally outperform index buy-and-hold and that the portfolio criteria can beat equal-weight index portfolios.
we backtested it — Sharpe -0.10 · quant relevance 0.98
The paper presents a framework for evaluating LLM-discovered trading strategies without look-ahead leakage and while accounting for the full set of searched candidates. In real-data backtests on US stocks and ETFs, passive benchmarks pass its tests, while every evaluated LLM-discovered strategy fails after out-of-sample and multiple-testing-aware evaluation.
140 new papers cleared the quant-relevance screen in this period. Open the Quant Paper 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.