ABSTRACT OF PAPER
Title: Econometric measurements of Abnormal Returns: A historical analysis from 1960s to 1980s
Author: Wang QingChao
In real world, traders invest in certain stocks whose returns tend to be abnormal through analysis of all data available, while the incentive lies in the probability of using abnormal returns to get extra profits. They may employ various data and information available at the moment, such as financial reports, stock splits, regulatory policy changes, etc. to analyse security returns. Correspondingly, in the empirical side of academic research, how to measure abnormal returns by applying sample data, information and estimating methods becomes an urgent issue. This paper first reviews different measurements of abnormal returns through model specification. The first section can be categorised into two subgroups: abnormal returns which are treated as error terms; abnormal returns which are viewed as structural factors. To be more specific, the former is primarily formalised in 1969 by Fama, Fisher, Jensen, Roll (henceforth FFJR) (1969) by employing a quasi-CAPM. In parallel with the development of model specification, different estimating methods are used in different models. This paper discusses various estimating methods for different model specifications in the second section. It is worthwhile to point out that the improvement of estimating methods seems to play a secondary role in comparison with the issue of model specification. A quite different picture is portrayed with respect to data. Data either take the role in estimating parameters in different models as discussed above, or directly measuring abnormal returns without any estimating models or methods, e.g. financial data that are employed in Value Line ranking system. In the third section of this paper, these two major roles of data will be explicitly discussed. In the fourth section, this paper will inspect two criteria in judging measures of abnormal returns. The first and foremost one is practical criterion—feasibility of using measures of abnormal returns to construct profitable trading strategies; the second one is statistical criterion—validity of statistical tests for different measures of abnormal returns. The last section of this paper intends to evaluate different roles that the above research orientations play in measuring abnormal returns between 1960s and 1980s. Specifically, with respect to two different criteria in judging measurements of abnormal returns, the advantages and disadvantages of every research area are explicitly discussed and contrasted. Furthermore, it is worthwhile to point out how these contrasts contribute to the history of econometrics.
Registred web users only can download this paper - Go back
Please note that files available for download have not been checked for viruses. These files have been submitted by authors of the conference to this web site. Conference organisers can't accept any responsibility for damages caused to users by downloading such files.


.jpg)