Login

Publications  •  Project Statistics

Glossary  •  Schools  •  Disciplines
People Search: 
   
Title/Abstract Search: 

Dissertation Information for William L. Seaver

NAME:
- William L. Seaver

DEGREE:
- Ph.D.

DISCIPLINE:
- Statistics

SCHOOL:
- Texas A&M University (USA) (1973)

ADVISORS:
- None

COMMITTEE MEMBERS:
- None

MPACT Status: Incomplete - Not_Inspected

Title: A MULTIVARIATE ANALYSIS OF THE FINANCIAL STRUCTURE OF SELECTED INDUSTRIES

Abstract: The major hypothesis of this dissertation is that an industry has a distinct financial structure which would justify the use of a unique subset of financial ratios to describe differences within an industry over time and between companies. A secondary consideration is to evaluate the application of present multivariate methods within financial analysis, especially in regard to variable selection.
A selection of 29 financial ratios which are most common to the literature, simplest to calculate, and the most meaningful to financial information users is drawn from the ratio categories of liquidity, activity, profitability, leverage, and miscellaneous. These ratios are computer for firms in the following six industries: crude-oil producing, textile apparel, chemicals, electronic components, auto parts and accessories, and retail-food. The data for the six selected industries are taken from the Standard & Pooer's Compustat Tapes. A screening of the data revealed that there were missing values, data irregularities, outliers, and unusual financial circumstances such as negative common equity which had to be resolved before the data were usable for analysis.
Various multivariate methods are utilized in analyzing each industry for key financial ratios. A principal component analysis identifies the number of linear combinations of ratios that accounted for most of the financial variability within each industry and provides insight into redundant ratios. After a selection of 13 to 17 ratios for each industry, a multivariate analysis of variance (MANOVA) discloses only moderate differences between years but tremendous differences between companies for each of the six industries. The discriminant functions related to the MANOVA are used to reveal the dimensionality of hypothesis differences and to provide insight as to which ratios contribute the most to discriminating among years and firms. The importance of individual ratios for differentiating among years and companies is also evaluated by examining the correlation of each ratio with the discriminant as well as by the factorization of the Wilks' A (the multivariate parallel of the univariate F-statistic in ANOVA for hypothesis testing).
The multivariate analysis of the six industries shows that liquidity is generally not important for distinguishing among firms within an industry and that one ratio each from the categories of activity, profitability, and leverage is usually adequate for financial analysis. A further investigation of four industries from the manufacturing sector also indicates that financial ratios important to an industry would not necessarily be good discriminators between industries from the same industrial classification.

MPACT Scores for William L. Seaver

A = 0
C = 1
A+C = 1
T = 0
G = 0
W = 0
TD = 0
TA = 0
calculated 2009-05-19 09:35:51

Advisors and Advisees Graph