Model Misspecification and Underdiversification


  • Raman Uppal,

  • Tan Wang

    Search for more papers by this author
    • London Business School and CEPR, and University of British Columbia, respectively. We are extremely grateful for detailed discussions with Bernard Dumas and Larry Epstein. We also acknowledge comments and suggestions from an anonymous referee, Suleyman Basak, Murray Carlson, Greg Connor, Adlai Fisher, Francisco Gomes, Tim Johnson, Burton Hollifield, Kai Li, Pascal Maenhout, Jun Pan, Sinan Sarpca, Stephen Schaefer, Luis Viceira, Hongjun Yan, Bill Ziemba, and seminar participants at Lancaster University, London Business School, MIT Sloan School of Management, University of Amsterdam, University of British Columbia, University of North Carolina, University of Warwick, the 2002 meetings of the AFA, SIRIF Conference on Behavioral Finance, and the Conference on Recent Developments in Securities Valuation and Risk Management at London School of Economics.


In this paper, we study intertemporal portfolio choice when an investor accounts explicitly for model misspecification. We develop a framework that allows for ambiguity about not just the joint distribution of returns for all stocks in the portfolio, but also for different levels of ambiguity for the marginal distribution of returns for any subset of these stocks. We find that when the overall ambiguity about the joint distribution of returns is high, then small differences in ambiguity for the marginal return distribution will result in a portfolio that is significantly underdiversified relative to the standard mean-variance portfolio.