TITLE:
Applied Investment Research and the CRSP Stock Market Database: Celebrating 60 Years of Financial Research
AUTHORS:
John B. Guerard Jr., Dimitrios Thomakos, Foteini Kyriazi, Bijan Beheshti
KEYWORDS:
Stock Prices and Returns, Relative Stocks and Bond Returns, Portfolio Selection, Robust Regression, Efficient Markets
JOURNAL NAME:
Journal of Mathematical Finance,
Vol.15 No.3,
August
25,
2025
ABSTRACT: The Center for Research in Security Prices (CRSP) database was funded and created to address the premier financial database for the academic and financial communities. At the time that the CRSP was funded in 1960, there were no large sets of databases for financial research. Additional questions were analyzed regarding the usefulness of daily stock returns in event studies. Academicians, such as James Lorie and Lawrence Fisher, the CRSP co-Directors, were concerned with relative stock and bond monthly returns, 1926-1964. William Sharpe, were presented his analysis of mutual fund returns at one of the first CRSP research meetings. Barr Rosenberg was an early prolific researcher of equity models and made numerous CRSP presentations in the 1970s and early-1980s. Harry Markowitz and his research group at Daiwa Securities experimented with CRSP data for monthly stock portfolio selection of equity portfolios, presenting at several CRSP research seminars in the early 1990s. In this article, we update the Markowitz Daiwa Securities articles presented at CRSP. Expected stock returns are a key input to portfolio selection. Stock selection models often use momentum, analysts’ expectations, and fundamental data. We briefly trace the early economic theory of profits and stock prices, hypothesized by Wesley Clair Mitchell, the first Director of Research at the National Bureau of Economic Research, the NBER, 1910-1946. The Mitchell emphasis on corporate profits on business cycles was followed by the fundamental investing framework of the Graham and Dodd Security Analysis approach to investing. The fundamental approach to investing has been enhanced over the past 60 years by modeling stock data on momentum, analysts’ expectations, and fundamental using CRSP data. This analysis specifically updates research of the past 30 years.