MULTIVARIATE TRANSVARIATION THEORY AMONG SEVERAL DISTRIBUTIONS AND ITS ECONOMIC APPLICATIONS.
Abstract
The report describes the theory of transvariation for multivariate distribution functions. The author has defined and deduced the parameters of transvariation; first, without any assumption of parametric distribution functions (nonparametric case); and second, under the assumption of multivariate normal distributions. The research continued with the study of the theory of transvariation among several (three or more) multivariate distribution functions. For the purpose of simplifying computation, the author has introduced a linear transformation that allows the application of bivariate transvariation theory to the transformed variables. The multivariate normal distribution is considered after the proof of two Lemmas regarding the distribution of a linear function of correlated normal random variables. For the case of more than two multivariate distributions, Gini's aggregative method is applied to simplify further the computations in the applications. The applications of this paper are in the field of comparative static economics. They affirm the fruitfulness of transvariation theory as a quantitative method in comparative statics (intertemporal and interspatial comparative analysis). (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1968
- Accession Number
- AD0676907
Entities
People
- Camilo Dagum
Organizations
- Princeton University