Uncertainty-Aware Small-Cap Portfolios with News, Macro, and Technical Signals
Summary
This study combines language-model sentiment from financial news with macroeconomic indicators and technical signals to construct portfolios of Russell 2000 stocks. Its uncertainty-aware approach separates predicted risk into aleatoric and epistemic components and feeds those estimates into the covariance matrix used by portfolio allocators. It compares three selection regimes: signals attributed to stock-specific moves, macro-driven moves, and cases where both channels agree.
The reported results favor the separate stock-specific and macro legs over requiring both signals to fire together. Macro-driven selections appear useful at a one-day horizon under low or moderate transaction costs, while high costs erase that edge; at a 40-day horizon, slower macro repricing is reported to favor the macro leg. The strongest conservative result is reported for a 40-day macro-trigger strategy using GPT-4o mini sentiment, a Student-t target, and risk parity, with a Sharpe ratio of 2.33 at 100 basis points in costs. The excerpt does not describe the full test design or establish that these results generalize beyond the evaluated stocks and assumptions.
Key ideas
- The portfolio pipeline incorporates separate aleatoric and epistemic risk estimates into its covariance matrix.
- It distinguishes stock-specific triggers from macro-driven triggers and their intersection.
- Separate stock-specific and macro legs usually outperform requiring both signal channels to agree.
- Macro triggers may capture fast spillovers at short horizons, but transaction costs can remove the advantage.
- The reported best conservative configuration uses a 40-day holding period and risk parity, though test details are limited.
Tags
Full text
# 2608.12283 # Large Language Model-Driven Small-Capitalization Trading: Integrating Financial News Sentiment, Macroeconomic Indicators, and Technical Signals Large language models can extract richer signals from financial news than fixed sentiment lexicons, and recent work has explored feeding such signals into portfolio construction. We study an uncertainty-aware construction that feeds model-predicted risk -- decomposed into aleatoric and epistemic components -- directly into the covariance matrix of portfolio allocators, rather than treating portfolio risk as fixed or adjusting only expected returns. We evaluate the pipeline on Russell 2000 equities under three stock-selection regimes: a pure-alpha trigger that isolates abnormal stock moves not explained by macro indicators, a pure-beta trigger that captures macro-indicator moves before the stock itself fires, and a beta trigger in which both channels agree. Across the full holding-period grid, the separated pure-alpha and pure-beta legs usually dominate the beta intersection on Sharpe and return. Two horizons are especially informative. At one day, pure beta can work under low and moderate transaction costs because it captures immediate lead-lag spillovers from liquid macro and sector indicators into exposed small-cap stocks, but this advantage disappears at 100 bps when turnover and microstructure noise dominate. At 40 days, pure beta works for a different reason: slower macro repricing overtakes the firm-specific pure-alpha channel. The strongest conservative row is pure beta with GPT-4o mini sentiment, a Student-t target, a 40-day holding period, and risk parity allocation, reaching Sharpe 2.33 at 100 bps. The results suggest that stock-selection regime and allocator choice matter at least as much as the sentiment model, and that separating firm-specific and macro-exposure triggers is more informative than requiring both to fire simultaneously.
Shown in full with attribution under the source's licence. Licence: abstract CC0
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.