A Study on the Biases of Management Earnings Forecasts

碩士 === 國立臺北大學 === 會計學系 === 88 === Management earnings forecasts are frequently released by the public companies listed on Taiwan Security Exchange (TSE). In this study, we investigate whether managers of the public companies, under the hypothesis of information asymmetries, use earnings forecasts wi...

Full description

Bibliographic Details
Main Authors: Chau-Hsi Hsu, 許晁熙
Other Authors: Chung-Huey Huang, Ph.D.
Format: Others
Language:zh-TW
Published: 2000
Online Access:http://ndltd.ncl.edu.tw/handle/13188655795666302805
Description
Summary:碩士 === 國立臺北大學 === 會計學系 === 88 === Management earnings forecasts are frequently released by the public companies listed on Taiwan Security Exchange (TSE). In this study, we investigate whether managers of the public companies, under the hypothesis of information asymmetries, use earnings forecasts with optimistic biases to influence the stock prices. We first take both voluntary and mandatory management earnings forecasts during 1993 to 1997 to examine the existence and the determinants of optimistic biases contained in the management earnings forecasts. Then we test the stock returns during the forecast announcement periods to show whether investors can identify the biases in the forecasts and adjust the stock prices accordingly. Our empirical results demonstrate that both voluntary and mandatory forecasts have optimistic biases. Besides, earnings predictability, debt ratio, predicting period, bull market, new stock offering, common stock turnover rate, and industry membership are influential factors of voluntary earnings forecasts biases. Earnings predictability, earnings variability, debt ratio, times of forecasting, common stock turnover rate, and industry membership are influential factors of mandatory earnings forecasts biases. Finally, we find that, during the forecast announcement periods, the cumulative mean abnormal returns (CAR) of voluntary management earnings forecasts with optimistic biases are smaller than those without optimistic biases. Furthermore, the significant association between CAR and forecast biases indicates that investors can see through the biases and effectively filter the biases in the stock prices.