Predictability of earnings, precision of information, and accuracy of earnings expectations.
Item
-
Title
-
Predictability of earnings, precision of information, and accuracy of earnings expectations.
-
Identifier
-
AAI9510727
-
identifier
-
9510727
-
Creator
-
Tang, Yabing.
-
Contributor
-
Adviser: Steven Lustgarten
-
Date
-
1994
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Business Administration, Accounting | Economics, Finance | Statistics
-
Abstract
-
This study develops a simple model to explain the relationships among predictability of earnings, precision of information and accuracy of earnings expectations. It also empirically tests the relationships predicted by the model using earnings expectations of two groups of market participants: financial analysts and investors.;For analysts, this study uses a sample of 133,599 analysts' forecasts for 807 firms obtained from the IBES database. The accuracy of investors' earnings expectations is proxied by the variance in price changes during earnings announcements. The test results indicate that analysts'/investors' earnings expectations are relatively more (less) accurate (compared to mechanic time series forecasts) when they possess more (less) precise public and/or private information and when future earnings are less (more) predictable. Thus, the study provides theoretical and empirical evidence suggesting that earnings predictability and the precision of information are fundamental determinants of the accuracy of earnings expectations.;The evidence also indicates that analysts and investors are able to learn each other's private information by observing public information signals. Analysts can incorporate information contained in price and trading volume movements. Investors, on the other hand, are able to utilize analysts' information through processing analysts' forecasts.;The study uses several proxies for precision of information: variance of price-based earnings forecast errors, standardized trading volumes, dispersion of analysts' forecasts, and incremental accuracy of analysts' forecasts over price-based forecasts. These proxies are developed on the basis of theoretical work, and provide more powerful explanatory power in empirical tests than firm size and its related variables which are used in prior studies. This research finds that size is only weakly associated with information precision and provides little incremental explanatory power to forecast accuracy after the information effect has been captured in the model.;To explain analysts' accuracy, previous studies suggest that analysts possess a timing advantage in addition to information advantage. The results of this study suggest that the timing of forecasts has no effect on the accuracy of forecasts after the information effect is controlled. This means that if the forecasters possess equally precise information about a firm, their forecasts will be equally accurate, regardless of when the forecasts are made.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
