Using statistical methods for marketing researches
This article deals with clustering analysis of cigarette markets in the Ukraine. The technique is illustrated by an example where the author groups different cigarette brands, present at the Ukrainian market, by two parameters into five homogeneous segments. The author uses the following formula for computation of the distance between multidimensional objects: NEWLINE\[NEWLINER_{A,B}=\left(\sum_{j=1}^{n}w_{j}(z_{Aj}- z_{Bj})^2\right)^{1/2},NEWLINE\]NEWLINE where \(w_{j},\;j=1,\ldots,n\) are weights, \(z_{ij}=(x_{ij}-\bar x_{j})/\sigma_{j}\), where \(x_{ij}\) is the initial value of dimension \(j\) for object \(i\), and NEWLINE\[NEWLINE\bar x_{j}=m^{-1}\sum_{i=1}^{m}x_{ij},\quad \sigma_{j}=\left(m^{-1}\sum_{j=1}^{m}(x_{ij}-\bar x_{j})^2\right)^{1/2}.NEWLINE\]
This page was built for publication: Using statistical methods for marketing researches
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2740099)
