Files in this item



application/pdf9712284.pdf (3MB)Restricted to U of Illinois
(no description provided)PDF


Title:The incremental predictive ability of individual financial analysts
Author(s):Giullian, Marc Andrew
Doctoral Committee Chair(s):Kleinmuntz, Don N.
Department / Program:Accountancy
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Business Administration, Accounting
Abstract:Financial analysts are among the most influential group of users of financial accounting information. Because the FASB has advocated usefulness as the "overriding criterion" (FASB, 1980, p.26) to judge accounting choices, accountants have a stake in understanding this important group of financial statement users. The majority of existing accounting research concerning financial analysts focuses on aggregated analysts' earnings forecasts rather than individual analysts' forecasts. Studies in accounting have documented the superiority of aggregated analysts' earnings forecasts relative to models. This is in contrast to the robust result from years of judgment/decision making (JDM) research that human predictions are inferior to statistical model predictions. Prior accounting studies have also documented that analysts exhibit optimism when forecasting earnings.
Humans can make a significant contribution to accurate forecasting in spite of cognitive limitations. Some skills people bring to bear are cue identification, rapid adaptability to environmental changes and the evaluation of qualitative factors. Although statistical models are not well-equipped to utilize qualitative factors and be adaptable, they do offer consistency and significant computational power. Thus, the strengths of humans and statistical models in forecasting are complementary.
This research documents the incremental predictive ability of both individual financial analysts and statistical models in forecasting earnings. It also provides evidence that both individual financial analysts' and statistical models' incremental predictive ability varies between industries. In addition, tests show a pessimistic bias for individual analysts, contrary to prior studies. Additional evidence is presented regarding forecast accuracy for four different forecast generation methods.
Issue Date:1996
Rights Information:Copyright 1996 Giullian, Marc Andrew
Date Available in IDEALS:2011-05-07
Identifier in Online Catalog:AAI9712284
OCLC Identifier:(UMI)AAI9712284

This item appears in the following Collection(s)

Item Statistics