We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
THE STATISTICAL VALIDITY OF THE RATIO METHOD IN FINANCIAL ANALYSIS: AN EMPIRICAL EXAMINATION: A REPLY.
- Authors
McDonald, Bill; Morris, Michael H.
- Abstract
This article presents a reply to a comment on the statistical validity of the ratio method in financial analysis. Certain comments address the issues of normality, proportionality and homoscedasticity in cross-sectional data, but he fails to appreciate the substantial differences which were found in the two subgroups of cross-sectional data. The original study provides empirical evidence on the behavior of a few commonly used ratios in both an intra-industry and cross-industry subsample. The ratio method assumes residual normality and as the Table 3 PRO model reveals, the presence of nonnormalities is substantially reduced in all cases. The ratio method assumes no intercept and no significant intercept for the PRO-A model of Table 3 was found in sill four cases. The follow-up study (McDonald and Morris, 1985) found significant nonlinearities, necessitating the use of a generalized functional form methodology to find the most appropriate transformation of the original data.
- Subjects
HETEROSCEDASTICITY; HOMOSCEDASTICITY; ANALYSIS of variance; ECONOMETRICS; RATIO analysis; CORPORATE finance
- Publication
Journal of Business Finance & Accounting, 1986, Vol 13, Issue 4, p633
- ISSN
0306-686X
- Publication type
Article
- DOI
10.1111/j.1468-5957.1986.tb00525.x