The Empirical Foundations of The Arbitrage Pricing Theory I: The Empirical Tests

The last thirty years have witnessed great changes in the field of finance associated with institutionally oriented students has been transformed into one dominated by scholars with amore scientific orientation. The Capital Asset Pricing Model (CAPM) was an outgrowth of this new emphasis and has served as the basis for modern portfolio theory for much of this period. The early tests of the CAPM (implemented with the equally-weighed index as a proxy for the market portfolio) revealed few violations of the zero beta model which could not be ascribed to statistical problems. In particular, the only reliably documented evidence concerned the slope of the empirical security market line which appeared shallower than that predicted by the theory. The tentative success of these initial investigations provided no hint of the disastrous empirical failure which followed: a large body of persuasive evidence which suggested that the dividends yields, market capitalizations, and price-earnings ratios of common stocks were strongly related to expected returns after risk adjustments based on the equally-weighted or value-weighted indices.