Abstract
This paper evaluates whether the primary and secondary dissemination of earnings forecast revisions by security analysts is reflected in security prices. Security prices were used to determine the profitability (before the cost of search) of trading strategies based on the nonpublic knowledge of forecast revisions. For a sample of 288 weekly earnings forecast revisions, the results were consistent with the hypothesis that early knowledge of forecast revisions could be used to form profitable trading strategies. Furthermore, the secondary dissemination of forecasts continued to have information content at the point of disclosure. These results are inconsistent with the strong form, but consistent with the semi-strong form, of market efficiency. Furthermore, the information contemporaneously available from public sources did not generate equivalently profitable trading rules, indicating that forecast revisions were not deducible from other publicly available information. Finally, some general public policy implications concerning mandatory disclosure of forecasts were drawn.
Original language | English (US) |
---|---|
Pages (from-to) | 661-680 |
Number of pages | 20 |
Journal | Accounting Review |
Volume | 57 |
Issue number | 4 |
State | Published - Oct 1982 |
Externally published | Yes |
Keywords
- Earnings forecasting
- Analytical forecasting
- Efficient markets
- Error rates
- Statistical forecasts
- Profit forecasting
- Random sampling
- Statistical significance
- Sampling errors
- Investors