| Аннотация | Improving population’s quality of life is a key goal of the state. In this regard, it is very important
to correctly measure its level and, accordingly, classify the country’s regions by quality of life indicators.
Most research in this area involves dividing variables into groups, unifying variables in each group and
building an integral indicator, grouping or clustering objects as a linear convolution of variables with
weights. Such approaches have their drawbacks due to the subjectivity of expert estimates, instability of
the coefficients of the main component, inability to work with ordinal data, etc. Thus, the purpose of
this study is to build a methodology for classifying the regions of the Russian Federation by quality of life
indicators devoid of the above disadvantages. The proposed method is based on the concept of Pareto
optimality well-known in Economics according to which all the regions are divided into disjoint classes.
After dividing variables into groups we recommend using Pareto class as a representative of the category
instead of the traditional unification and construction of intra-group convolutions, which is obtained
after the intra-group Pareto classification, and building the final Pareto classification of the regions of the
Russian Federation on the basis of the obtained intra-group Pareto classes. The advantage of the proposed
approach is that it can be applied on the ordinal data, that is, when some variables are characterized only
by their order and there are no exact values for each region. In addition, the algorithm is undemanding
for computing power and does not use expert estimates, except for the selection of research variables. The main results of the study are the construction of a classification of the Russian Federation regions
by quality of life indicators, comparison with traditional approaches and analysis of the features of the
proposed methodology. |
|---|
| Ключевые слова | regional ranking, population’s quality of life indicators, stratification, Pareto ratio, Pareto dominance, Pareto classification, Pareto optimum, quality of life. |
|---|
