An Information Interpretation of Financial Analyst Superiority in Forecasting Earnings Export

@article{citeulike:3326603, abstract = {A model is presented that provides determinants of ex ante forecast accuracy and examines conditions under which a particular forecasting approach (i.e., the use of analysts' forecasts) would yield a superior measure for earnings surprise. Insights are given on how to partition observations using ex ante measures of forecast accuracy for comparing alternative surrogates for earnings surprise. The determinants of financial analysts' forecast (FAF) superiority are modeled in the context of the information environment of the firm. FAF superiority can be defined as the ratio of the variance of a random-walk forecasting error to the variance of FAF forecasting error. It is shown theoretically that FAF superiority is positively related to the dimensionality of the information set used to generate FAF. This suggests that using a time-series model rather than FAF to generate a measure of the market's expectation of earnings induces measurement error that is positively related to the measure of information dimensionality.}, author = {Brown and Lawrence, D. and Richardson and Gordon, D. and Schwager and Steven, J.}, citeulike-article-id = {3326603}, journal = {Journal of Accounting Research}, keywords = {accuracy, analysis, analysts, earnings, financial, forecasting, models, regression, series, time, variable}, number = {1}, pages = {49--68}, posted-at = {2008-09-24 01:42:39}, priority = {2}, title = {An Information Interpretation of Financial Analyst Superiority in Forecasting Earnings}, volume = {25}, year = {1987} }

TY - JOUR ID - citeulike:3326603 L3 - citeulike-article-id:3326603 N2 - A model is presented that provides determinants of ex ante forecast accuracy and examines conditions under which a particular forecasting approach (i.e., the use of analysts' forecasts) would yield a superior measure for earnings surprise. Insights are given on how to partition observations using ex ante measures of forecast accuracy for comparing alternative surrogates for earnings surprise. The determinants of financial analysts' forecast (FAF) superiority are modeled in the context of the information environment of the firm. FAF superiority can be defined as the ratio of the variance of a random-walk forecasting error to the variance of FAF forecasting error. It is shown theoretically that FAF superiority is positively related to the dimensionality of the information set used to generate FAF. This suggests that using a time-series model rather than FAF to generate a measure of the market's expectation of earnings induces measurement error that is positively related to the measure of information dimensionality. IS - 1 JF - Journal of Accounting Research EP - 68 TI - An Information Interpretation of Financial Analyst Superiority in Forecasting Earnings VL - 25 SP - 49 KW - accuracy KW - analysis KW - analysts KW - earnings KW - financial KW - forecasting KW - models KW - regression KW - series KW - time KW - variable AU - Brown AU - Lawrence, D AU - Richardson AU - Gordon, D AU - Schwager AU - Steven, J PY - 1987/// ER -