Date of Award

1987

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Accounting

Abstract

The usefulness of current cost (CC) information, as compared to historical cost (HC) information, has been a focus of debate over the last decade. Inconsistent research findings concerning this question suggest that additional evidence is needed. Given the intensive merger activity in recent years and the significance of mergers to the economy and shareholders, this study empirically investigated the usefulness of current cost information in merger prediction. Fifty-seven firms that were merged between 1981 and 1984 were selected from the FASB databank, and an equal number of nonmerged firms were matched with the merged firms. Using the subsample for 1981-82, linear discriminant models were developed from six sets of variables which had been selected based on the findings of earlier studies and related merger theories. The models were applied to the second subsample for 1983-84 to test their predictive value. The discriminant analysis results show that CC information classified mergers as effectively as HC information. CC information also added incremental classificatory power to the HC information. However, in terms of prediction errors, the results of pairwise tests indicate that HC information outperformed its CC counterpart, which also did not add any incremental power to the prediction. The findings strongly suggest that care be exercised in the application of a successful classificatory model.

Pages

182

DOI

10.31390/gradschool_disstheses.4381

Share

COinS