Multivariate Statistical Techniques

This paper serves as an overview of various multivariate statistical techniques that can be used to analyse and describe survey datasets. Such analyses are useful for gaining a better understanding of results and the interpretation thereof. In particular this paper evaluates some of the standard statistical techniques that can be used to analyse consumption patterns of households, both in terms of expenditure on individual goods or baskets of goods. The ultimate aim is to determine whether these techniques – e.g. multivariate tests of significance and cluster analysis techniques – can be successfully employed to assist in forming representative household groups in terms of expenditure patterns. As such this paper is exploratory and does not aim to obtain final results. It is envisaged that more comprehensive research will be done at a later stage of the project. Initial findings suggest that there is merit in using multivariate statistical techniques to form representative household groups, although certain constraints and problems may be encountered.

Issue Date:
Publication Type:
Working or Discussion Paper
PURL Identifier:
Total Pages:
Series Statement:
Background Paper 2003: 1

 Record created 2017-04-01, last modified 2018-01-22

Download fulltext

Rate this document:

Rate this document:
(Not yet reviewed)