Change in Systematic Bias and News as Driver for Analyst Forecast Revision - A New Forecast Revision Model

2009 ◽  
Author(s):  
Sebastian Gell ◽  
Carsten Homburg ◽  
Katja Schulze
Keyword(s):  
Analyst Forecast ◽  
Systematic Bias ◽  
Forecast Revision

Stock Splits: Tests of the Earnings Signaling and Attention Directing Hypotheses Using Analyst Forecasts and Revisions

1994 ◽  
Vol 9 (3) ◽  
pp. 411-422 ◽  
Author(s):  
David T. Doran
Keyword(s):  
Earnings Forecasts ◽  
Analyst Forecasts ◽  
Analyst Forecast ◽  
Specific Level ◽  
Future Earnings ◽  
Earnings Information ◽  
Stock Split ◽  
Forecast Revision ◽  
Earnings Performance ◽  
Signaling Hypothesis

The major findings of this study are: (1) earnings performance of splitting firms is favorable relative to preevent longterm analyst (Value Line) forecasts; (2) analysts significantly revise earnings forecasts upward in response to stock split announcements; and (3) in the case of stock split announcing firms, there is a high correlation between future earnings performance and analyst forecast revision. These findings indicate that stock split announcements convey “permanent” earnings information to the market, and security analysts scrutinize the earnings signal at the firm specific level. The results support both the earnings signaling hypothesis and the attention directing hypothesis concerning stock split events.

Discussion: “Dispersion in Analysts' Earnings Forecasts as a Measure of Uncertainty”

1998 ◽  
Vol 13 (3) ◽  
pp. 271-274 ◽  
Author(s):  
Lawrence D. Brown
Keyword(s):  
Stock Price ◽  
Annual Report ◽  
Earnings Forecasts ◽  
Analyst Forecast ◽  
Earnings Forecast ◽  
Data Set ◽  
Forecast Dispersion ◽  
Forecast Revision ◽  
Analyst Earnings ◽  
The Impact

This paper tackles an interesting question; namely, whether dispersion in analysts' earnings forecasts reflects uncertainty about firms' future economic performance. It improves on the extant literature in three ways. First, it uses detailed analyst earnings forecast data to estimate analyst forecast dispersion and revision. The contrasting evidence of Morse, Stephan, and Stice (1991) and Brown and Han (1992), who respectively used consensus and detailed analyst data to examine the impact of earnings announcements on forecast dispersion, suggest that detailed data are preferable for determining the data set on which analysts' forecasts are conditioned. Second, it relates forecast dispersion to both analyst earnings forecast revision and stock price reaction to the subsequent earnings announcement. Previous studies related forecast dispersion to either analyst forecast revision (e.g., Stickel 1989) or to subsequent stock price movements (e.g., Daley et al. [1988]), but not to both revision and returns. Third, it includes the interim quarters along with the annual report. In contrast, previous research focused on the annual report, ignoring the interims (Daley et al. [1988]).

Factors affecting an analyst forecast revision— Taiwan and the United States: A comparison

1997 ◽  
Vol 32 (1) ◽  
pp. 63-77 ◽  
Author(s):  
Glen D. Moyes ◽  
Kyungjoo Park ◽  
Andrew Minglong Wang ◽  
Patricia A. Williams
Keyword(s):  
United States ◽  
The United States ◽  
Analyst Forecast ◽  
Factors Affecting ◽  
Forecast Revision

How Completely Do Analysts Incorporate Firm-Specific and Industry Information in Their Forecasts? Evidence and Implications for Post-Forecast Revision Drift

2021 ◽  
pp. 0148558X2110437
Author(s):  
Sami Keskek ◽  
Senyo Tse
Keyword(s):  
Analyst Forecast ◽  
Positive Relation ◽  
Forecast Revisions ◽  
Forecast Revision ◽  
Available Information

Prior studies find a positive relation between analyst forecast revisions and upcoming news, suggesting that analysts’ forecast revisions are incomplete with respect to available information. In this study, we use the association between forecast revisions and upcoming news to measure forecast completeness and show that post-forecast-revision drift is higher when forecasts are incomplete. We follow Hui and Yeung’s (2013) approach to separate forecast revision news into industry-wide and firm-specific components because they find that drift is primarily associated with the industry component. We find that forecast revisions are less complete for industry-wide news than for firm-specific news. Furthermore, analysts’ industry-wide revisions are less complete early in the year and when the underlying news is bad, and we find stronger post-forecast-revision drift in those cases. We also show that analysts who were optimistic in prior periods tend to issue forecasts that are less complete and that generate stronger drift than forecasts by other analysts. Our findings provide an explanation for the drift that contrasts with prior studies that attribute the drift to investors’ slow assimilation of the news in forecast revisions. Thus, our study sheds light on analysts’ role in conveying firm-specific and industry-wide news to investors and on the implications for post-forecast-revision drift.

Counter culture: Causes, extent and solutions of systematic bias in the analysis of behavioural counts

2019 ◽  
Author(s):  
Joel L Pick ◽  
Nyil Khwaja ◽  
Michael A. Spence ◽  
Malika Ihle ◽  
Shinichi Nakagawa
Keyword(s):  
Recent Literature ◽  
Null Model ◽  
Mechanistic Explanation ◽  
Effect Sizes ◽  
Systematic Bias ◽  
Intra Class Correlation ◽  
Counter Culture ◽  
Provisioning Rate ◽  
Parental Provisioning ◽  
Observation Period

We often quantify a behaviour by counting the number of times it occurs within a specific, short observation period. Measuring behaviour in such a way is typically unavoidable but induces error. This error acts to systematically reduce effect sizes, including metrics of particular interest to behavioural and evolutionary ecologists such as R2, repeatability (intra-class correlation, ICC) and heritability. Through introducing a null model, the Poisson process, for modelling the frequency of behaviour, we give a mechanistic explanation of how this problem arises and demonstrate how it makes comparisons between studies and species problematic, because the magnitude of the error depends on how frequently the behaviour has been observed (e.g. as a function of the observation period) as well as how biologically variable the behaviour is. Importantly, the degree of error is predictable and so can be corrected for. Using the example of parental provisioning rate in birds, we assess the applicability of our null model for modelling the frequency of behaviour. We then review recent literature and demonstrate that the error is rarely accounted for in current analyses. We highlight the problems that arise from this and provide solutions. We further discuss the biological implications of deviations from our null model, and highlight the new avenues of research that they may provide. Adopting our recommendations into analyses of behavioural counts will improve the accuracy of estimated effect sizes and allow meaningful comparisons to be made between studies.

Sign in / Sign up

Export Citation Format

Share Document