Multivariate analysis of marketing data - applications for bricolage market

• Author: Fanaru M.
• Pages: 419-428

Bulletin of Transilvania University of Braşov

Series V: Economic Sciences • Vol. 9 (58) No. 2 – 2016

Multivariate analysis of marketing data -

applications for bricolage market

Mihai FÂNARU1

Abstract: By using concepts and analytical tools for computing, marketing is directly

related to the quantitative methods of economic research and other areas where the

efficiency of systems performances are studied. Any activity of the company must be

programmed and carried out taking into account the consumer. Providing a complete

success in business requires the entrepreneur to see the company and its products through

the consumers eyes, to act as representative of its clients in order to acquire and satisfy their

desires. Through its complex specific activities, marketing aims to provide goods and

services the consumers require or right merchandise in the right quantity at the right price at

the right time and place. An important consideration in capturing the link between marketing

and multivariate statistical analysis is that it provides more powerful instruments that allow

researchers to discover relationships between multiple configurations of the relationship

between variables, configurations that would otherwise remain hidden or barely visible. In

addition, most methods can do this with good accuracy, with the possibility of testing the

statistical significance by calculating the level of confidence associated with the link

validation to the entire population and not just the investigated sample.

Key-words: marketing, consumer, multivariate statistical analysis, variables

1. Introduction

A good analogy to express the relationship between the information contribution of

univariate and bivariate methods on one hand and that of the multivariate, on the

other hand, could be for example, just like making a comparison between

black&white and color photos. As it can be understood from its name, multivariate

means, in the simplest sense, ”many variables considered simultaneously” in the

analysis, in most cases involving more than two variables. In this respect,

multivariate methods can analyze the networking of several variables analyzed

simultaneously, based on a specific model for each individual method.

Most of the techniques identify configurations of similarities or links between

variables, or make predictions or estimate the relative importance of each variable in

predicting or explaining certain variables.

1 Transilvania University of Braşov, fanaru.mihai.cosmin@unitbv.ro