Principal Component Analysis - A Powerful Tool in Computing Marketing Information

• Author: Constantin, C.
• Position: Faculty of Economic Sciences and Business Administration, Transilvania University of Brasov
• Pages: 25-30

Bulletin of the Transilvania University of Braşov

Series V: Economic Sciences • Vol. 7 (56) No. 2 - 2014

PRINCIPAL COMPONENT ANALYSIS -

A POWERFUL TOOL IN COMPUTING

MARKETING INFORMATION

Cristinel CONSTANTIN1

Abstract: This paper is about an instrumental research regarding a

powerful multivariate data analysis method which can be used by the

researchers in order to obtain valuable information for decision makers that

need to solve the marketing problem a company face with. The literature

stresses the need to avoid the multicollinearity phenomenon in multivariate

analysis and the features of Principal Component An alysis (PCA) in reducing

the number of variables that could be correlated with each other to a small

number of principal components that are uncorrelated. In this respect, the

paper presents step-by-step the process of applying the PCA in marketing

research when we use a large number of variables that naturally are

collinear.

Key words: multivariate analysis, multicollinearity, principal component

analysis, marketing research.

1 Faculty of Economic Sciences and Business Administration, Transilvania University of Braşov.

1. Introduction

In the most cases of marketing research the

descriptive analysis and the univariate or

bivariate inferential analyses are not enough

for obtaining that information needed by the

decision factors that face with a marketing

problem and order such a research. The

multivariate analyses extract the main

information from a large number of variables

and offer additional details that can support

the decision process. The computation of

such methods is quite complicated but the

modern information systems can assist the

researchers to obtain the best information.

Nevertheless the correct using of the

multivariate methods and the results

interpretation are very important. In this

respect, the present research aims to assist

mainly the young researchers in using the

Principal Component Analysis (PCA) as

one of the most popular multivariate data

analysis methods. The theoreticians and

practitioners can also benefit from a detailed

description of the PCA applying on a certain

set of data.

2. Literature review

Principal component analysis (PCA) is a

method of data processing consisting in the

extraction of a small number of synthetic

variables, called principal components,

from a large number of variables measured

in order to explain a certain phenomenon.

Principal components are a sequence of

projections of the data, mutually

uncorrelated and ordered in variance,