Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.250804
Title: An empirical analysis of takeover predictions in the UK : application of artificial neural networks and logistic regression
Author: Yuzbasioglu, Asim
ISNI:       0000 0001 3576 3903
Awarding Body: University of Plymouth
Current Institution: University of Plymouth
Date of Award: 2002
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
This study undertakes an empirical analysis of takeover predictions in the UK. The objectives of this research are twofold. First, whether it is possible to predict or identify takeover targets before they receive any takeover bid. Second, to test whether it is possible to improve prediction outcome by extending firm specific characteristics such as corporate governance variables as well as employing a different technique that has started becoming an established analytical tool by its extensive application in corporate finance field. In order to test the first objective, Logistic Regression (LR) and Artificial Neural Networks (ANNs) have been applied as modelling techniques for predicting target companies in the UK. Hence by applying ANNs in takeover predictions, their prediction ability in target classification is tested and results are compared to the LR results. For the second objective, in addition to the company financial variables, non-financial characteristics, corporate governance characteristics, of companies are employed. For the fist time, ANNs are applied to corporate governance variables in takeover prediction purposes. In the final section, two groups of variables are combined to test whether the previous outcomes of financial and non-financial variables could be improved. However the results suggest that predicting takeovers, by employing publicly available information that is already reflected in the share price of the companies, is not likely at least by employing current techniques of LR and ANNs. These results are consistent with the semi-strong form of the efficient market hypothesis.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.250804  DOI: Not available
Keywords: Statistics Mathematical statistics Operations research Artificial intelligence
Share: