Institutional Equity Holdings Prediction Using Node Affinities of Dynamic Graphs

2026-07-13

AI Summary

This study addresses the challenge of predicting institutional investors’ future equity holdings, which is hindered by reporting lags, disclosure noise, and behavioral persistence. Framing the problem as node affinity prediction on a dynamic bipartite graph, this work introduces the first affinity-based benchmark for institutional ownership forecasting and proposes the Node Affinity prediction model with Virtual states (NAVIS), integrating dynamic graph representation learning within the Temporal Graph Benchmark (TGB) framework. Experimental results on real-world 13F filings demonstrate that temporal and structural signals alone capture the majority of predictable information, with node features providing limited additional value. NAVIS achieves a test NDCG of 0.9127, substantially outperforming existing dynamic graph models and heuristic baselines.

Abstract

Institutional equity holdings disclosed in SEC Form 13F filings provide a rich temporal record of portfolio decisions by large investment managers. However, forecasting future allocations and modeling future demand remains challenging due to disclosure lags, reporting noise, and strong persistence in institutional behavior. We introduce the first benchmark for these tasks using temporal graph machine learning, framing holdings prediction as node affinity prediction -- i.e., forecasting portfolio weights -- on a discrete-time temporal bipartite graph of managers and securities extracted from preprocessed filings. On a sampled dataset comprising 99 managers and the S\&P 500 index (503 securities, 209,351 temporal edges across 48 quarters from 2013--2025), Node Affinity prediction model using Virtual State (NAVIS) achieves a state-of-the-art test Normalized Discounted Cumulative Gain (NDCG) of 0.9127 with features (0.9121 without), outperforming all dynamic graph representation learning competitors by a substantial margin, and outperforming all heuristic methods. Remarkably, a simple Exponential Moving Average baseline achieves 0.8882, surpassing all dynamic graph models except NAVIS and all heuristics except Persistent Forecast (0.8891), highlighting the strong smoothness and persistence of institutional portfolios. Domain-specific node features provide only marginal gains (<1.2\%), indicating that temporal and structural signals in the 13F ownership graph already capture most of the predictable information. By benchmarking a suite of Temporal Graph Benchmark (TGB) models under the node affinity prediction setting, both with and without features, on real-world 13F data, this work provides a reproducible foundation for temporal graph machine learning in holdings prediction and portfolio allocation.

Problem

• institutional equity holdings
• 13F filings
• holdings prediction
• temporal graph
• portfolio allocation

Innovation

• node affinity prediction
• temporal graph machine learning
• institutional equity holdings
• dynamic bipartite graph
• NAVIS