building techniques to publicly available information could have permitted an investor to earn a portfolio return in excess of the return which was commensurate with the portfolio risk. The question of equity market efficiency over time is an area of constant disagreement, especially between practitioners and theoreticians. The disagreement is really a matter of degree. Practitioners believe there is enough inefficiency in price adjustment to offer the diligent analyst the opportunity of excess returns net of transactions costs. Academicians maintain that it is precisely the actions of these astute analysts that cause consistent trading opportunities to be non-existent. This issue can only be answered empirically and it is hoped that this study will bring new evidence to bear. The study undertakes an extensive statistical investigation of aggregate quarterly stock prices (the Standard and Poors Index of Five Hundred Common Stocks) and their relationship to a leading indicator of business activity.1 The Box-Jenkins methodology is utilized to build a transfer function model relating changes in the National Bureau of Economic Research Leading Composite Index to subsequent stock price changes. The model is tested in a fifty quarter holdout sample and found to be successful at forecasting stock price changes one quarter ahead. The study covers the period 1948 to 1974. Model parameters are estimated in the first half of the period and tested in the second. The model is evaluated with respect to predicted R 2, hit rate for forecasts of "up" and "down" markets, and accumulated wealth of a stock market/treasury bills trading rule, compared to a buy and hold strategy for various levels of transactions costs. The following conclusions are drawn from this research: (1) stock prices move in