Timely Aggregate Analyst Forecasts as Better Proxies for Market Earnings Expectations

Previous research (e.g., O'Brien [1988], Stickel [1990], and Brown [1991]) has documented that timely composites of analysts' forecasts are superior to the mean forecast in terms of predictive ability. An alternative criterion in choosing an earnings expectation proxy is market association, whereby the forecast whose error is most highly correlated with abnormal returns is the proxy of choice (Foster [1977]). This paper shows that timely composites are superior to the mean on the market association dimension. The results are robust to the three timely composites considered by Brown [1991] and pertain to each of five years and two deflators