Skip to content
All library documents

Modeling Investor Demand with a Factor-Based Asset Pricing System

Article BigQuant

Summary

This article argues that asset prices and expected returns can be understood by modeling who demands assets, rather than focusing only on how company characteristics predict returns. With supply often relatively stable, changes in investor demand can move clearing prices and shape future returns. The article explains the Koijen–Yogo framework: derive portfolio weights from investors’ utility maximization, then represent expected excess returns and factor loadings through asset characteristics, yielding a factor structure for demand.

It describes empirical work using institutional holdings reports and standard stock characteristics. The reported findings attribute more cross-sectional return variation to demand-side forces than to supply changes, and suggest that smaller institutions and households can explain more crisis-era cross-sectional variation than the largest institutions. A demand-based estimate that assumes long-run mean reversion in latent demand is also reported to predict future returns. These are summaries of cited research, not a ready-to-trade strategy; the evidence is focused mainly on U.S. equities and institutional holdings, with measurement limits for investors whose holdings are not reported.

Key ideas

  • Asset demand can influence prices when investors’ trades are large relative to available supply.
  • The Koijen–Yogo framework derives optimal investor weights and imposes factor structure on expected returns and factor loadings.
  • Institutional holdings data can help estimate how investor groups respond to asset characteristics.
  • The cited research finds demand-side forces explain more return variation than supply-side changes.
  • Smaller institutions and households may contribute more to cross-sectional price variation than the largest institutions.
  • An estimate based on mean-reverting latent demand is reported to predict future returns, though coverage is incomplete.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.