Search for a command to run...
Alfred Cowles was an investment counselor in Colorado Springs, Colorado. He told me in 1952 that after the stock market crash of 1929, he realized that he did not understand the workings of the economy, and so in 1931 he stopped publishing his market advisory letter, and began research on stock market forecasting.2 Harold T. Davis, a mathematician who spent summers in Colorado Springs, put him in touch with Irving Fisher at Yale, president of the fledgling Econometric Society. Fisher had known Cowles' father and uncle when all three were undergraduates at Yale. Davis and Fisher proposed an economics research organization and a journal. The idea appealed to Cowles, and he agreed to provide financing. The Cowles Commission was founded in Colorado Springs in 1932, and Econometrica began publishing in 1933. The Commission's Articles of In1 Two earlier retrospective works on the Cowles Commission are Christ (1952), a nonmathematical history of its first twenty years, and Clifford Hildreth (1986), a detailed and technical account of its work at the University of Chicago from 1939 to 1955. Theodore W. Anderson (1991) and James Heckman (1992) discuss Trygve Haavelmo's contributions. Christ (1985, Section VII), deals with the identification problem. Roy Epstein (1987)and Mary Morgan (1990) discuss the history of econometrics in general, including the Cowles Commission's contributions. 2 Cowles and his associates generated random stock market advice, and random portfolio choices, and compared them with actual market newsletters and actual fire insurance company portfolios. They found that on the average the actual advice and portfolios were slightly inferior to the random ones. The best actual advice and portfolios were about as good as the best random ones, but the worst actual ones were worse than the worst random ones (Cowles 1933).