STRAPSim: A Portfolio Similarity Metric for ETF Alignment and Portfolio Trades

2025-09-28

AI Summary

Existing portfolio similarity measures (e.g., Jaccard) rely on asset overlap or static weights, failing to capture semantic associations, partial overlaps, and heterogeneous weight distributions. To address this, we propose STRAPSim—a semantic-aware, two-level residual-aware similarity metric. STRAPSim jointly models component-level semantic matching (inspired by BERTScore), dynamically reweights constituent assets, and employs a residual-aware greedy alignment mechanism; final similarity is computed via a two-stage weighted aggregation for robustness. Evaluated on ETF recommendation, portfolio trading, and risk alignment tasks, STRAPSim consistently outperforms baselines—including Jaccard and weighted Jaccard—achieving state-of-the-art Spearman correlation with realized return similarity. It effectively overcomes the expressiveness limitations of conventional static metrics.

Technology Category

• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Search and Optimization: Sampling/Simulation-based Search
• Multiagent Systems: Agent-Based Simulation and Emergent Behavior

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success

Abstract

Accurately measuring portfolio similarity is critical for a wide range of financial applications, including Exchange-traded Fund (ETF) recommendation, portfolio trading, and risk alignment. Existing similarity measures often rely on exact asset overlap or static distance metrics, which fail to capture similarities among the constituents (e.g., securities within the portfolio) as well as nuanced relationships between partially overlapping portfolios with heterogeneous weights. We introduce STRAPSim (Semantic, Two-level, Residual-Aware Portfolio Similarity), a novel method that computes portfolio similarity by matching constituents based on semantic similarity, weighting them according to their portfolio share, and aggregating results via residual-aware greedy alignment. We benchmark our approach against Jaccard, weighted Jaccard, as well as BERTScore-inspired variants across public classification, regression, and recommendation tasks, as well as on corporate bond ETF datasets. Empirical results show that our method consistently outperforms baselines in predictive accuracy and ranking alignment, achieving the highest Spearman correlation with return-based similarity. By leveraging constituent-aware matching and dynamic reweighting, portfolio similarity offers a scalable, interpretable framework for comparing structured asset baskets, demonstrating its utility in ETF benchmarking, portfolio construction, and systematic execution.

Problem

• Measures portfolio similarity for ETF alignment and trading
• Captures nuanced relationships between partially overlapping portfolios
• Improves predictive accuracy in financial classification and recommendation

Innovation

• Uses semantic similarity for constituent matching
• Applies residual-aware greedy alignment for aggregation
• Leverages dynamic reweighting based on portfolio shares