Розроблення алгоритму методики експрес-оцінки рівня та якості життя населення регіону
The article presents the rapid assessment methodology that allows analysts to carry out qualitative monitoring of the living standards of the population using a wide range of methods for statistics processing (factor analysis, cluster analysis, discriminant analysis, method of combining indicators o...
Збережено в:
| Дата: | 2021 |
|---|---|
| Автори: | , , |
| Формат: | Стаття |
| Мова: | Англійська |
| Опубліковано: |
The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute"
2021
|
| Теми: | |
| Онлайн доступ: | https://journal.iasa.kpi.ua/article/view/239871 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | System research and information technologies |
| Завантажити файл: |  |
Репозитарії
System research and information technologies| _version_ | 1866302756118069248 |
|---|---|
| author | Lesnichaya, Marina Kolchina, Oksana Pahomov, Evgenii |
| author_facet | Lesnichaya, Marina Kolchina, Oksana Pahomov, Evgenii |
| author_sort | Lesnichaya, Marina |
| baseUrl_str | http://journal.iasa.kpi.ua/oai |
| collection | OJS |
| datestamp_date | 2021-09-16T11:48:22Z |
| description | The article presents the rapid assessment methodology that allows analysts to carry out qualitative monitoring of the living standards of the population using a wide range of methods for statistics processing (factor analysis, cluster analysis, discriminant analysis, method of combining indicators of different dimensions). This methodology is characterized by the high speed of mathematical calculations, availability to users with different skill levels and universal applicability to various study objects. The rapid assessment method is intended for screening the living standards of the population and activity quality of the territorial authorities, taking into account a different set of indicators. This article offers the author’s indicator system for assessing the living standards and quality of life of the population. The methodology algorithm describes flowcharts of the index method for combining statistical observations of different dimensions, which make it possible to automate the process of territory ranking. The study covers 12 urban districts and 43 municipal districts of Rostov Oblast. The methodology described in the article will help eliminate a subjective factor while monitoring, rationally distribute financial resources allocated annually by the authorities to support programs for socio-economic development of the territory, increase the economic efficiency and implementation speed of innovative projects that have a direct impact on the living standards and quality of life of the population. |
| doi_str_mv | 10.20535/SRIT.2308-8893.2021.2.04 |
| first_indexed | 2025-07-17T10:27:26Z |
| format | Article |
| fulltext |
M.A. Lesnichaya, O.A. Kolchina, E.V. Pahomov, 2021
50 ISSN 1681–6048 System Research & Information Technologies, 2021, № 2
UDC 332.145
DOI: 10.20535/SRIT.2308-8893.2021.2.04
DEVELOPING AN ALGORITHM FOR RAPID ASSESSMENT
OF LIVING STANDARDS AND QUALITY OF LIFE
OF THE POPULATION IN THE REGION
M.A. LESNICHAYA, O.A. KOLCHINA, E.V. PAHOMOV
Abstract. The article presents the rapid assessment methodology that allows analysts
to carry out qualitative monitoring of the living standards of the population using a
wide range of methods for statistics processing (factor analysis, cluster analysis, dis-
criminant analysis, method of combining indicators of different dimensions). This
methodology is characterized by the high speed of mathematical calculations, avail-
ability to users with different skill levels and universal applicability to various study
objects. The rapid assessment method is intended for screening the living standards of
the population and activity quality of the territorial authorities, taking into account a
different set of indicators. This article offers the author’s indicator system for assessing
the living standards and quality of life of the population. The methodology algorithm
describes flowcharts of the index method for combining statistical observations of dif-
ferent dimensions, which make it possible to automate the process of territory ranking.
The study covers 12 urban districts and 43 municipal districts of Rostov Oblast. The
methodology described in the article will help eliminate a subjective factor while
monitoring, rationally distribute financial resources allocated annually by the authori-
ties to support programs for socio-economic development of the territory, increase the
economic efficiency and implementation speed of innovative projects that have a direct
impact on the living standards and quality of life of the population.
Keywords: living standards, variable reduction, clustering, discriminant analysis,
ranking.
INTRODUCTION
According to the Strategy of Spatial Development of the Russian Federation for
the period up to 2025 (approved by the Order of the Government of the Russian
Federation dated February 13, 2019 No. 207-р), the priority development trend is
to ensure the sustainable and balanced spatial development of the Russian Federa-
tion aimed at reducing interregional differences in the living standards and quality
of life of the population, accelerating economic growth and technological devel-
opment, and ensuring the national security of the country. In order to achieve this
goal, the authorities need to direct their management measures at the solution of
long-standing and most urgent problems connected with social inequality and
high levels of poverty. A real decrease in these indicators is the main criterion for
socio-economic progress of the Russian Federation and its subjects.
In this regard, it becomes necessary to develop and implement a scien-
tifically grounded concept of improving the living standards and quality of life.
This concept, first of all, should be based on an effective system of mathematical
assessment methods, allowing to form an information and analytical platform for
supporting the management decisions based on various data sources. This article
tries to achieve the objective by developing a universal algorithm for rapid
Developing an algorithm for rapid assessment of living standards and quality of life
Системні дослідження та інформаційні технології, 2021, № 2 51
assessment intended to screen the living standards of the population and quality of
the territorial authorities’ activities, taking into account a different set of
indicators and system analysis methods. 12 urban districts and 43 municipal dis-
tricts of Rostov Oblast were selected as a study object, because Rostov Oblast has
implemented a successful practice of “Building a regional poverty profile” aimed
at identifying poor citizens (the practice was supported by Lipetsk Oblast, the Re-
public of Sakha, Novgorod Oblast) [https://smarteka.com/practices/ vyavlenie-
maloimusih-grazdan-s-maksimal-nojdekompoziciej-harakteristik-maloimusih.].
The proposed rapid assessment method will enhance this service prospects
accumulating the data not only by category of poor citizens, but also on a number
of key indicators (investments, level of consumption possibilities, level of digi-
talization, number of middle class representatives, etc.) influencing on increasing
the living standards and quality of life of the population.
BLOCK DIAGRAM OF RAPID ASSESSMENT OF THE LIVING STANDARDS
AND QUALITY OF LIFE OF THE POPULATION
Modern methods for assessing the living standards of the population refer to the
complex systems which are difficult to be formalized. They are mainly based on
an expert approach, inaccurate description and measurement errors. These all
significantly distort the study results. The developed method for rapid assessment
has a number of advantages, because it accumulates various system analysis
methods complementing each other and allowing to avoid the data assessment
subjectivity. In the context of global digitalization this method provides a
decision-maker with an objective information base for the implementation of
strategic decisions in a short period of time.
FORMATION OF INDICATOR SYSTEM
The first stage of rapid assessment deals with development of an indicator system
based on collection, systematization, analysis of quantitative and qualitative in-
formation at the level of empirical indicators. The indicator system should be as
complete as possible, taking into account all input and output elements to avoid
distortion in the relationships between the model parameters. As data processing
is carried out in the information system with a complex of system analysis meth-
ods, the number of data can be maximum, because the obtained results will be
interpreted in an easier way and errors will be minimal. Due to the fact that many
factors influence on the living standards and quality of life of the population, and
depending on a time lag, some of them may have a random nature, it is more ap-
propriate to consider this impact through stochastic processes, which will reveal a
trend in the assessment indicators.
When forming an author’s indicator system of the living standards and qual-
ity of life in the municipalities of Rostov Oblast, the work of N.V. Zubarevich
(Faculty of Geography, Moscow State University) “Assessment of the quality of
life index” was used, the analysis of the methodology of the British research cen-
ter The Economist Intelligence Unit (analysis division of The Economist), The
Legatum Prosperity Index (LPI) and the City Development Index (CDI) [N.V.
Zubarevich 2019] were also taken into account. The analyzed methods and indi-
ces have their indisputable advantages. However, they have a major disadvantage,
M.A. Lesnichaya, O.A. Kolchina, E.V. Pahomov
ISSN 1681–6048 System Research & Information Technologies, 2021, № 2 52
namely, they have no indicators reflecting the development and implementation
level of digital technologies, which have a significant impact on the living stan-
dards and quality of life of the population (Fig. 1). During the pandemic, these are
digital technologies that made it possible for people to continue working,
communicate with their loved ones and provide themselves with essential goods.
The digital technologies allowed people who have temporarily lost their jobs to
learn the skills of remote work and new professions, allowed small businesses to
increase their profits by using online platforms with an unlimited number of con-
sumers, etc. The proposed indicator system meets modern realities and new
challenges. The author’s indicator system is presented in Table 1.
Start
Input of initial data
Using the existing scorecard
Development of the author's system of
indicators
Construction of regression equations characterizing the dependence of the levels of the series on the time
variable by the method of analytical alignment
Stage 1
Stage 2
Yes
Normalization of indicators using the Z-matrix of factor analysis
Reducing the number of variables (data reduction) using the method of factor analysis
Determination of the structure of relationships between variables, identification of latent indicators and the
degree of influence of variables on the object of research by the method of principal components (factor
analysis)
Yes
Yes
Classification of territorial entities according to the standard of living of the population, using
agglomerative and iterative methods of cluster analysis
Factor 1 Factor 2 Factor 3 Factor n
No No No
Cluster 1
N
o
No
Cluster 2 Cluster 3 Cluster n
No NoNo
Assessment of the quality of classification by the method of discriminant analysis
Territories with a
high standard of
living
Territories with an
average standard of
living
Areas with a low
standard of living
….
Yes Yes Yes Yes
Ranking of territorial entities by the standard of living of the population
End
Stage 3
Fig. 1. Algorithm for the author’s assessment of the living standards and quality of life of
the population
Developing an algorithm for rapid assessment of living standards and quality of life
Системні дослідження та інформаційні технології, 2021, № 2 53
T a b l e 1 . The author’s indicator system for assessing the living standards and
quality of life of the population
Income of the population Human potential
Average monthly salary of one employee.
Unequal degree in distribution of income
of the population.
Poverty level of the population.
Consumption possibilities of the population.
Savings rate
Number of the young adults.
Education level.
Life expectancy.
Air pollutant emissions
Socio-economic indicators in the territory Digitalization
Migration inflow (+), migration outflow (-).
Local budget surplus (+), local budget
deficit (-).
Crime rate.
Level of registered unemployment.
Investment amount
Number of legal and other persons, who
have a key of an enhanced qualified
electronic signature.
Rendering of public and municipal
services through Multifunctional Public
Services Centers (MPSCs).
Proportion of the population using the
Internet
For a deeper analysis of statistical data, the rapid assessment methods
accumulate indicators over several years to build up time series using the
analytic method of fitting constants, as the change in levels of time series is
expressed by certain mathematical functions and the rules are set in such a
way that the series value is predicted quite accurately. This improves the
quality of the study results.
GROUPING OF VARIABLES AND CLASSIFICATION OF THE STUDY
OBJECTS
Grouping the diverse data into homogeneous populations classified according to
the certain rules, taking into account their characteristic features. The first stage of
assessment involves the developing an indicator system and identifying depend-
encies between the indicators. As many elements, which give a complete picture
of the study object, influence on the living standards and quality of life of the
population, they must be aggregated into groups of indicators which will strongly
correlate with each other. In order to achieve our objective, the second stage of
assessment involves the factor analysis, which is a set of sequential actions aimed
at accumulating indicators that form a set, called factors. The factor analysis is
used to minimize the number of independent values, detect correlation signs and
latent (hidden) indicators, and allows to normalize the data, that is necessary for
further research. The factor analysis put variables into a linear set of F-factors
with minimum errors for the best representation of X :
p
k
jKjkj UFaX
1
,
where kF , ,2,1k are latent indicators, p are common factors, ajk values are
called factor loadings, and jU , mj ,2,1 , are specific factors.
M.A. Lesnichaya, O.A. Kolchina, E.V. Pahomov
ISSN 1681–6048 System Research & Information Technologies, 2021, № 2 54
The object for transformation in factor analysis is a variance-covariance ma-
trix, consisting of X data array. The result of these data compression is a reduced
correlation matrix, i.e. the matrix is decreased in dimension, but not the data [Ful-
vio Castellacci 2018, J. Langford 2005]. The factor analysis conducted by the
principal component method will result in the cumulative variance presented in
Table 2 and in the rotated component matrix obtained by rotation method: Vari-
max rotation with Kaiser normalization shown in Table 3.
To implement the principal component analysis, it is necessary to normalize
the indicators upon standard deviations. The Z matrix based on the initial data is
defined as follows:
j
jij
ij S
xx
Z
,
where jS is a standard deviation, mj ,,2,1 ; ni ,,2,1 , jx is an average
value of the j-th feature to normalize the indicators is necessary for the further
research.
T a b l e 2 . Explained variance
Initial eigenvalues
Extraction sums
of squared loadings
Rotation sums
of squared loadings
C
om
p
on
en
t
Total
Variance,
%
Cumula-
tive, %
Total
Variance,
%
Cumula-
tive, %
Total
Variance,
%
Cumula-
tive, %
1 4,241 41,375 47,124 4,241 41,375 47,124 3,658 40,642 40,642
2 1,975 15,579 69,066 1,975 15,579 69,066 2,146 23,849 64,492
3 1,559 12,357 86,393 1,559 12,357 86,393 1,971 21,901 86,393
4 1,452 11,464 75,856 1,379 11,464 75,856 1,379 19,257 75,856
5 0,852 9,464 55,856
6 ,,, ,,, ,,, ,,, ,,, ,,, ,,, ,,, ,,,
The values exceeding one indicate that highly correlated indicators are
classified upon four factors (indicators of the total variance F1 “Population in-
come” — 41,375%, F2 “Socio-economic indicators” — 15,579%, F3 “Human
potential” — 12,357%, F4 “Digitalization” — 11,464%).
The factor analysis results (the rotated Z matrix generated by rotation
method: Varimax rotation with Kaiser normalization) are shown in Table 2,
which confirms that the indicators are classified upon four factors.
T a b l e 3 . Results of rotated matrix (generated by rotation method: Varimax
rotation with Kaiser normalization)
Factor
Coefficient name
F1 F2 F3 F4
Poverty level of the population (VAR00003) 0,851 0,093 0,351 0,258
Unemployment rate (VAR00009) 0,063 0,728 0,289 0,059
Level of education (VAR00012) -0,133 0,041 0,909 0,123
Proportion of the population using the Internet 0,231 0,089 0,153 0,658
… … … …
Developing an algorithm for rapid assessment of living standards and quality of life
Системні дослідження та інформаційні технології, 2021, № 2 55
The Table shows that the poverty level of the population strongly correlates
with the first factor, its factor loading is 0,851, the unemployment rate tends to the
second factor with the factor loading of 0,728, the education level indicator has
the factor loading of 0,909, which clearly indicates a correlation with the third
group of factors, as the loading on other groups is very low, the proportion of the
population using the Internet will be referred to the fourth factor with the loading
of 0,658.
Another important feature of the factor analysis is that factor loadings indi-
cate not only belonging to a particular group, but also reflect the impact of indica-
tors as a whole on the system or factors. This is graphically represented in the fol-
lowing form (Fig. 2).
This advantage makes the factor analysis more attractive increasing the study
objectivity, because there is no need to use an expert approach. This fact is also im-
portant when assessing the living standards and quality of life of the population.
The third stage of the algorithm involves the classification of municipalities
according to the living standards of the population. Rostov Oblast is characterized
by a high degree of differentiation in the development of its municipalities, there-
fore, the regional authorities need to solve a difficult task, i.e. to remove the exist-
ing differences not only in terms of socio-economic development, but also in the
living standards of the population. For these purposes, first of all, they are to have
a mathematical apparatus and verified system of indicators (the first and second
stages of the rapid assessment), which make it possible to qualitatively classify
the territory into groups close in the development indicators. In our opinion, the
clustering methods, which are included in rapid assessment, namely, the iterative
k-means method and seven hierarchical agglomerative methods deal well with
this task.
The cluster analysis is a mechanism consisting of a set of methods and vari-
ous similarity measures allowing to classify the variables mXXX ,,, 21 . The
main task of cluster analysis is to aggregate the groups of similar indicators,
called clusters (clumps, groups, bundles). Past research experience allows us to
conclude that the iterative k-means method gives the qualitative classification re-
sults. Its disadvantage is that a decision-maker needs to set a number of clusters
on his own, and the hierarchical agglomerative methods do not require this
[I.D. Mandel 1988, S.A. Ayvazyan 1989, T. Hastie, R. Tibshirani, J. Friedman
2009]. The cluster analysis results are visually presented either in scatter diagrams
or in dendrograms, Fig. 3.
Unemploy
ment rate
Fig. 2. Influence of variables or factors including factor loading
M.A. Lesnichaya, O.A. Kolchina, E.V. Pahomov
ISSN 1681–6048 System Research & Information Technologies, 2021, № 2 56
Therefore, you may use all hierarchical agglomerative methods with all
similarity measures to determine the number of clusters and their structural
content, and then carry out the iterative k-means clustering, although the
procedure is cumbersome, but with minimum errors.
As an example, the adjacency matrix with results of clustering of municipali-
ties in Rostov Oblast is presented below. The calculations have shown that the
most effective of the agglomerative metric classifiers are the “intragroup connec-
tions” and “Ward’s method”. The Table shows that classifying into five clusters is
an optimal solution. The numbers indicate the administrative territorial units of
Rostov Oblast [M.A. Lesnichaya, O.A. Kolchina, H.A. Kondzharyan 2019].
T a b l e 4 . Adjacency matrix of clustering results
Distance
M
et
h
od
s
Euclidean
distance
Squared
Euclidean
distance
Block distance
(Manhattan
distance)
Minkowski
distance …
C
on
ne
ct
io
ns
a
m
on
g
gr
ou
ps
Contagion
effect
Contagion
effect
1) 10, 48, 22, 35,
47, 18, 13, 40, 53,
29, 30,7,23,31
2) 28, 39, 8, 25, 3,
2, 44, 46, 54, 38,
26, 51, 34, 45, 50,
52, 55
3) 1,20
4)15,27,49,41,14,42
,21,11,
1)10,48,22,35,47,18,
38,54,
24,30,7,29,13,40,53,
26,34,51,55,31
2)8,39,25,28,3,45,50,
52,2,44,46
3)1,20
4)15,27,49,41,1
…
In
tr
ag
ro
up
c
on
ne
ct
io
ns
1)10,48,22,41,35,
47,18,38,54,51,55
2)8,39,25,28,3,45
,50,52,2,44,46
3)17
4)1,2026,34,13,4
0,53,7,29,30,24,3
1
1)10,48,22,41,35,4
7,18,38,54,31
2)13,40,53,7,29,
30,24
3)26,34,51,55,1,20
4)8,39,25,28,3,45,
50,52,2,44,46
1)10,48,22,41,35,47
,18,38,54,31,13,40,
53,7,29,30,24,26,34
,51,55
2)1,20,8,39,25,28,3,
45,50,52,2,44,46
3)15,27,49,12,14,42
,
4) 5,23,36
1)10,48,22,35,47,18,
38,54,31
2)13,40,29,7,30,53,2
4,26,34,51
3)16,43,9,19,6,4
4)14,42,21,32,15,49,
27,41,11,33,12,37
…
… … … … … …
Fig. 3. Cluster analysis results presented in graphical form
5,0
4,5
4,0
3,5
3,0
2,5
2,0
1,5
S
e
p
a
l W
ig
h
t
Iris Dateset, KMeans clustering with k=3
1 –
2 –
3 –
1
2
3
Petal Lenght
D
is
ta
nc
e
Example of a dendrogram
Subject of inquiry
Developing an algorithm for rapid assessment of living standards and quality of life
Системні дослідження та інформаційні технології, 2021, № 2 57
The iterative k-means clustering results are presented in Table 5.
Table 5. K-means clustering results
Observation
number
Cluster Distance
Observation
number
Cluster Distance
1 1 12,615 29 4 4,632
2 2 6,305 30 5 4,995
3 1 4,486 31 4 9,098
4 3 8,575 32 3 5,365
5 2 4,515 33 3 4,798
… … … … … …
The k-means method is convenient and very easy to interpret the results. The
Table shows that each observation is assigned its own cluster number depending
on the distance, then an expert should structurally fill the groups. Observations
present a study object. In our case, these are the municipalities of Rostov Oblast.
The invaluable advantage of k-means method is the classification quality [Roberta
Comunian and Lauren England 2019]. However, even such a voluminous classifi-
cation procedure, including all agglomerative and iterative methods, does not ex-
clude measurement errors, therefore, at the next stage, it is necessary to assess the
resulting typology.
CLASSIFICATION QUALITY ASSESSMENT
It is advisable to use the discriminant analysis with Mahalanobis distance and
Wilks lambda while assessing the obtained clustering results. This method has
revealed the municipalities that were gathered into a different cluster by mistake.
If we consider the discriminant constructing algorithm, it will represent the identi-
fication of a variable that causes the greatest differences among the observations
[I.D. Mandel 1988, S.A. Ayvazyan 1989, T. Hastie, R. Tibshirani, J. Friedman
2009]. The results of discriminant analysis of the municipalities in Rostov Oblast
are presented in Table. 6.
T a b l e 6 . Discriminant analysis results
Senior group
Case
number
Factor
group Predicted
group
P(D>d | G=g) P(G=g | D=d)
Squared Mahalanobis
distance
to the centroid
p df
1 1 1 0,873 3 0,825 1,378
2 1 1** 0,628 3 0,753 4,521
3 2 2 0,542 3 0,513 1,617
4 3 2 0,758 3 0,423 2,251
… … … … … … …
In the sample for each group of observations, the point position is deter-
mined in multidimensional space. This point is an average value for all variables,
M.A. Lesnichaya, O.A. Kolchina, E.V. Pahomov
ISSN 1681–6048 System Research & Information Technologies, 2021, № 2 58
i.e. centroid groups. Then, for each observation, the squared Mahalanobis distance
from a group centroid is calculated. The closer the observation is to the set, the
more minimum the Mahalanobis distance is to it (the observation is classified cor-
rectly). The “Predicted group” column (Table 6) has a mark “**”, indicating an
incorrectly assigned variable to the set. The results of discriminant analysis can
also be presented in a classification matrix containing the number of variables
located on the matrix diagonal, i.e. correctly classified and those that do not be-
long to their classes. When conducting the research, assessing the quality of the
resulting classification is necessary, because the observations represent the spe-
cific municipalities, groups of people, who, due to measurement errors, may re-
ceive less financial assistance, therefore any errors in this area are critical and un-
acceptable. The clustering results on the living standards of the population of
Rostov Oblast with the use of discriminant analysis are presented in Table. 7.
T a b l e 7. Results of clustering of municipalities in Rostov Oblast
Clusters Municipalities of Rostov Oblast
Cluster 1
Cities: Rostov-on-Don (53), Volgodonsk (46), Taganrog (54), Gukovo (47),
Zverevo (49)
Cluster 2
Cities: Azov (44), Bataysk (45), Shakhty (56), Donetsk (48),
Novocherkassk (51), Kamensk-Shakhtinsky (50)
Cluster 3
Aksaysky (2), Ust-Donetsky (39) Districts, Kamensk-Shakhtinsky (50),
Remontnensky (32), Proletarsky (31), Neklinovsky (26), Tatsinsky (38)
Districts, Novoshakhtinsk (52), Morozovsky (24), Volgodonskoy (8),
Bagayevsky (3), Oktyabrsky (28), Vesyolovsky (7), Krasnosulinsky
Districts (18), Zimovnikovsky District (13)
Cluster 4
Belokalitvinsky (4), Verkhnedonskoy (6), Dubovsky (9), Zavetinsky (11),
Kasharsky (16), Kuybyshevsky (19), Rodionovo-Nesvetaysky (33),
Sholokhovsky (43), Martynovsky (20), Azovsky (1 ) Yegorlyksky District (10),
Zernogradsky District (12), Kagalnitsky (14), Kamensky (15),
Matveyevo-Kurgansky (21), Millerovsky (22), Oblivsky (27), Tsimlyansky (41),
Chertkovsky (42) Districts , Orlovsky (29), Salsky (34) Districts, Peschano-
kopsky (30), Semikarakorsky (35), Myasnikovsky (25), Tselinsky
Districts (40)
Cluster 5 Bokovsky District (5), Milyutinsky District (23), Sovetsky District (36),
Tarasovsky District (37), Konstantinovsky District (17)
The municipalities of Rostov Oblast were accumulated in 5 clusters with dif-
ferent living standards of the population from a higher (cluster I) to a lower one
(cluster V), that makes it possible for the authorities to distribute financial re-
sources more efficiently, support various categories of the population and develop
programs for improving the living standards, taking into account the specifics of
the territories included in different clusters.
RANKING OF TERRITORIAL ENTITIES
The final stage of rapid assessment is to determine a place of the studied territory
in terms of the living standards of the population in the overall ranking of ad-
ministrative-territorial entities using the index method of ranking statistical ob-
servations [M.A. Lyamina 2008]. This algorithm is presented in the flowchart (a
program in the VBA language is developed) in Fig. 4.
Developing an algorithm for rapid assessment of living standards and quality of life
Системні дослідження та інформаційні технології, 2021, № 2 59
The method essence is to form an equal range scale of real values of the
investigated statistical observations from a minimum indicator to a maximum one.
Each range is assigned a rank which corresponds to the interval number. By
means of a graduation scale, the ranks are summed up, forming an integral
indicator. This, on the one hand, will allow to include the maximum set of
variables in the assessment system, and, on the other hand, based on the author's
indicator system, to make a scientifically grounded rating of administrative-
territorial entities.
Ranking is an important stage of monitoring, as it helps evaluate the effec-
tive activities of the authorities both in the field of improving the living standards
of the population and in other areas, using a rather simple mechanism.
DISCUSSION
According to the UN and other public organizations, Russia lags significantly be-
hind the leading countries of the world in various integral indicators of quality of
life. Therefore, the problems of studying the characteristics of the living standards
and quality of life of the population are significant in modern economic theory,
and it becomes especially urgent to develop an effective tool for their
comprehensive assessment.
A theoretical study of the quality of life was begun in the 1860s in
economically developed countries. A significant contribution to defining the
Fig. 4. Flowcharts of the index method algorithm for ranking statistical observations
M.A. Lesnichaya, O.A. Kolchina, E.V. Pahomov
ISSN 1681–6048 System Research & Information Technologies, 2021, № 2 60
“quality of life” category as a multifaceted concept of modern social progress was
made by famous foreign scientists, such as A. Toynbee, D. Bell, J. Galbraith, R.
Aron, H. Kahn, A. Viner, H. Marcuse, and E. Fromm.
The economic aspects of the living standards and quality of life of the
population are considered in the works of M.V. Borschevsky, L.A. Velikhov,
M.V. Glazyrin, B.S. Zhikharevich, I.A. Ilyin, A.E. Koguta, V.N. Leksin, N.V.
Lvov, V.Ya. Lyubovny, V.F. Mayer, V.E. Rokhchin, V.M. Rutgeiser, A.E.
Surinov and others. Social problems are studied in the works of N.A. Aitov, E.G.
Ayayimitsy, A.V. Baranov, A.V. Dmitriev, E.N. Zaborova, M.N. Mezhevich,
A.E. Polovinkin, I.I. Sigov and others.
In recent years the problems of economic and statistical assessment of the
living standards and quality of life have been actively discussed by the domestic
economists, such as S.A. Ayvazyan, A.M. Dubrov, V.M. Zherebin, V.S. Knyazevsky,
V.S. Mkhitaryan, L.I. Nivorozhkina, V.D. Raitsin, N.M. Rimashevskaya, A.N.
Romanov, I.I. Eliseeva, L.I. Troshin, G.N. Khubaev and others.
The issues of overcoming unevenness and stabilizing significant imbalances
in the socio-economic development levels of the Russian regions and the
problems of forming a single economic space are considered in the works of A.G.
Granberg, E.B. Kibalov, V.I. Klistorin, V.V. Kuleshov, A.M. Lavrov, B.L.
Lavrovsky, V.N. Leksin, V.Yu. Malov, V.D. Marshak, V.E. Seliverstov, V.I.
Suslov, S.A. Suspitsyn, V.A. Shabashev and others.
Despite the high relevance of intraregional differentiation, Russia still
neglects its removal. The authors study certain aspects of this phenomenon: the
issues of socio-economic development of municipalities (E.G. Animitsa, R.V.
Babun, G.Yu. Vetrov, A.G. Voronin, A.P. Gaponenko, B.S. Zhikharevich, V.V.
Ivanov, E.A. Kartaeva, Yu.S. Kolesnikov, V.A. Lapin, V.N. Leksin, Z.Z.
Mullagaleeva, P.A. Orekhovsky, A.N. Shvetsov and others) and local government
(A.A. Vasiliev, A.N. Shirokov, S.N. Yurkova).
Much attention is paid to the study of sustainable development problems in
the works of such domestic authors as I.Ya. Blekhtsin, S.N. Bobylev and V.I.
Vernadsky, I.F. Gazizullin, A.A. Golub, V.I. Danilov-Danilyan, V.A. Koptyug,
N.N. Moiseev, O.S. Pchelintsev, A.D. Ursul and others. Among the foreign
researchers of these problems are D. Bell, D. Meadows and J. Forrester.
MATERIALS AND METHODS
The statistical data published in monthly reports of the Rostov Regional State
Statistics Committee, statistical yearbooks and reference books have formed the
information and empirical base of the study.
The theoretical basis of the study consists of the fundamental concepts of the
management theory of socio-economic systems, works of domestic and foreign
scientists in the field of studying the living standards and quality of life of the
population, uneven spatial economic development of the regions, digitalization of
the regional and municipal economy, organization and management in the public
and state sectors, service economy, institutional economics and the legislative
regulatory documents of the government authorities and local government bodies
of the Russian Federation.
The study methodology is based on:
the system analysis methods and mathematical modelling used for design-
ing a concept and modelling the sustainable development of the region in the con-
ditions of inter-municipal differentiation;
Developing an algorithm for rapid assessment of living standards and quality of life
Системні дослідження та інформаційні технології, 2021, № 2 61
the factor analysis methods used for determining the dominant indicators
of development of the regions and municipalities;
the cluster analysis methods used for solving the problems of cluster con-
vergence analysis of socio-economic development levels of the regions and dif-
ferentiation of municipalities;
the economic dynamics forecasting methods used for predicting the dy-
namics of regional development;
the management theory methods used for creating management models
for integrated sustainable development of municipalities;
the discriminant analysis method used for assessing the quality of territory
classification and identifying the factors contributed the most in intergroup differ-
ences (focusing on the values of Wilks Lambda and squared Mahalanobis dis-
tance)[ W.R. Klecka 1989].
In their study the authors have used domestic and foreign experience of
economic and mathematical tools for analyzing and predicting the living
standards and quality of life of the population. The study was carried out using
general software (Microsoft Excel) and special software (STATISTICA, SPSS).
CONCLUSIONS
The proposed rapid assessment algorithm will form the basis for an analytical
program designed to support the management decisions made by the authorities to
improve the living standards and quality of life in order to remove inter-municipal
differences.
The management tasks that can be solved by the proposed assessment
method are as follows:
to simulate various scenarios for the situation development, which is an
important criterion in the dynamically changing conditions of modern economy;
by means of ranking, to assess the effective actions of the authorities;
to implement the experience of more successful administrative-territorial
entities in their own territory;
to reveal the development differences of the territories, identify the indica-
tors that have the greatest impact on stratification of the population in terms of the
living standards, and to reveal the systemic problems decreasing the living stan-
dards of the specific territories;
to allow the authorities rationally to distribute financial resources aimed at
supporting the specific territories in order to improve the quality, speed and effec-
tiveness of the proposed development programs.
The developed system for rapid assessment of the living standards and
quality of life of the population is a basic screening tool for supporting
management decisions. Due to a diverse set of methods for processing statistical
data, the methodology demonstrates high quality and speed of calculations, the
versatility of assessment allows users with different skill levels to conduct
researches, changing a system of variables and object to meet their requirements,
and rationally to allocate financial resources aimed at improving the living
standards and quality of life population. The proposed methodology will improve
the existing service in Rostov Oblast “Building a regional poverty profile”,
which is currently aimed only at providing assistance to poor citizens
[https://smarteka.com/practices/vyavlenie-maloimusih-grazdan-s-maksimal-
nojdekompoziciej-harakteristik-maloimusih.].
M.A. Lesnichaya, O.A. Kolchina, E.V. Pahomov
ISSN 1681–6048 System Research & Information Technologies, 2021, № 2 62
The proposed algorithm efficiency and effectiveness was proved by the state
statistics data of the municipalities of Rostov Oblast.
Author contributions: Writing—review & editing, M.L., O.K. and E.P. All
authors have read and agreed to the published version of the manuscript.
Funding: The study was financially supported by the Russian Foundation
for Basic Research (RFBR), the project “Developing a concept of improving the
living standards and quality of life of the population in the region in the condi-
tions of inter-municipal differentiation, taking into account the economy digitali-
zation” No. 20-010-00815 A.
Conflicts of interest: The authors declare no conflict of interest.
REFERENCES
1. W.R. Klecka, Factor, discriminant and cluster analysis, (in Russian). Moscow: Fi-
nance and Statistics, 1989, 215 p.
2. I.D. Mandel, Cluster analysis, (in Russian). Moscow: Finance and Statistics, 1988, 176 p.
3. S.A. Ayvazyan, V.M. Buchstaber, I.S. Enyukov, and L.D. Meshalkin, Applied statis-
tics: classification and dimensionality reduction, (in Russian). Moscow: Finance
and Statistics, 1989.
4. T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, 2nd
edition. Springer, 2009, 533 p.
5. J. Langford, Tutorial on Practical Prediction Theory for Classification, 2005, 28 p.
6. M. Zhambyu, Hierarchical cluster analysis and correspondence, (in Russian). Mos-
cow: Finance and Statistics, 1988, 345 p.
7. N.V. Zubarevich, “Russian regions in January-July 2019: Mix of growth, stagnation
and slump”, Russian Economic Development, vol. 26, no. 10.
8. Roberta Comunian and Lauren England, “Creative Clusters and the Evolution of Knowl-
edge and Skills: From Industrial to Creative Glassmaking”, Geoforum, 99, pp. 238–247,
2019.
9. M.A. Lesnichaya, O.A. Kolchina, and H.A. Kondzharyan, “Elaboration of a concep-
tual scheme of municipal units classification according to the level of budgetary pro-
vision on the base of cluster analysis (2019)”, Proceedings of the 2019 IEEE Con-
ference of Russian Young Researchers in Electrical and Electronic Engineering,
ElConRus 2019, article no. 8657216, pp. 1415–1420.
10. Marcelo Cano-Kollmann, John Cantwell, Thomas J. Hannigan, Ram Mudambi, and
Jaeyong Song, “Knowledge Connectivity: An Agenda for Innovation Research in Inter-
national Business”, Journal of International Business Studies, 47, pp. 255–262, 2016.
11. Fulvio Castellacci, “Co-Evolutionary Growth: A System Dynamics Model”, Eco-
nomic Modelling, 70, pp. 272–287, 2018.
12. M.A. Lyamina, “An approach to design of intelligent system rapid assessment for
the mortgage climate of the constituent entities of the Russian Federation” (in Rus-
sian), Technical science, no. 10 (87), pp. 23–28, 2008.
13. Smarteka.com. Available: https://smarteka.com/practices/vyavlenie-maloimusih-
grazdan-s-maksimal-nojdekompoziciej-harakteristik-maloimusih.
14. The Economist Intelligence Unit’s quality-of-life index. Available: https://www.
eiu.com/n/ (accessed on 26.06.20).
15. United Nations Development Programme, Human Development Reports 2019.
Available: http://hdr.undp.org/en/content/2019-human-development-index-ranking
(accessed on 26.06.20).
Received 18.04.2021
INFORMATION ON THE ARTICLE
Marina A. Lesnichaya, ORCID: 0000-0001-8978-5104, Institute of Management in Eco-
nomic, Environmental and Social Systems, Southern Federal University, Russia, e-mail:
grezinam@yandex.ru
Developing an algorithm for rapid assessment of living standards and quality of life
Системні дослідження та інформаційні технології, 2021, № 2 63
Oksana A. Kolchina, ORCID: 0000-0003-4161-9769, Institute of Management in Eco-
nomic, Environmental and Social Systems, Southern Federal University, Russia, e-mail:
kolchinaoksana@gmail.com
Evgenii V. Pahomov, ORCID: 0000-0002-3898-1444, Institute of Management in Eco-
nomic, Environmental and Social Systems, Southern Federal University, Russia, e-mail:
evpahomov@sfedu.ru
РОЗРОБЛЕННЯ АЛГОРИТМУ МЕТОДИКИ ЕКСПРЕС-ОЦІНКИ РІВНЯ ТА
ЯКОСТІ ЖИТТЯ НАСЕЛЕННЯ РЕГІОНУ / М.О. Ліснича, О.О. Колчина,
E.В. Пахомов
Анотація. Запропоновано методику експрес-оцінки, яка дозволить аналітикам
проводити якісний моніторинг рівня життя населення, використовуючи широ-
кий набір методів оброблення статистики (факторний аналіз, кластерний
аналіз, дискримінантний аналіз, метод об’єднання різновимірних показників),
що відрізняється високою швидкістю математичних обчислень, доступністю
користувачам з різним рівнем кваліфікації, універсальністю застосування для
різних об’єктів дослідження. Метод експрес-оцінки призначено для скринінгу
рівня життя населення і якості діяльності органів влади територіальних утво-
рень з урахуванням набору різних показників. Запропоновано авторську сис-
тему показників оцінювання рівня та якості життя населення. В алгоритмі ме-
тодики описано блок-схеми індексного методу об’єднання різновимірних
статистичних спостережень, що дозволяють автоматизувати процес ранжуван-
ня територій. Як об’єкт дослідження обрано 12 міських і 43 муніципальних
округів району Ростовської області. Застосування описаної методики дасть
змогу: вилучити суб’єктивний фактор під час проведення моніторингу;
раціонально розподіляти фінансові ресурси, які щорічно виділяються органами
влади на підтримання програм соціально-економічного розвитку території;
підвищити економічну ефективність і пришвидшити впровадження
інноваційних проєктів, що безпосередньо впливають на рівень і якість життя
населення.
Ключові слова: рівень життя населення, тренди динаміки, редукція змінних,
кластеризація, оцінювання якості класифікації, ранжування.
РАЗРАБОТКА АЛГОРИТМА МЕТОДИКИ ЭКСПРЕСС-ОЦЕНКИ УРОВНЯ И
КАЧЕСТВА ЖИЗНИ НАСЕЛЕНИЯ РЕГИОНА / М.А. Лесничая, О.А. Колчина,
Е.В. Пахомов
Аннотация. Предложена методика экспресс-оценки, позволяющая аналитикам
проводить качественный мониторинг уровня жизни населения, используя ши-
рокий набор методов обработки статистики (факторный анализ, кластерный
анализ, дискриминантный анализ, метод объединения разномерных показате-
лей), отличающаяся высокой скоростью математических вычислений, доступ-
ностью пользователям с разным уровнем квалификации, универсальностью
применения для различных объектов исследования. Метод экспресс-оценки
предназначен для скрининга уровня жизни населения и качества деятельности
органов власти территориальных образований, учитывающий различный на-
бор показателей. Предложена авторская система показателей оценки уровня и
качества жизни населения. В алгоритме методики описаны блок-схемы ин-
дексного метода объединения разномерных статистических наблюдений, по-
зволяющие автоматизировать процесс ранжирования территорий. В качестве
объекта исследования были выбраны 12 городских округов и 43 муниципаль-
ных района Ростовской области. Применение описанной методики позволит:
исключить субъективный фактор при проведении мониторинга; рационально
распределять финансовые ресурсы, ежегодно выделяемые органами власти на
поддержку программ социально-экономического развития территории; повы-
сить экономическую эффективность и скорость внедрения инновационных
проектов, которые оказывают непосредственное влияние на уровень и ка-
чество жизни населения.
Ключевые слова: уровень жизни населения, тренды динамики, редукция пе-
ременных, кластеризация, оценка качества классификации, ранжирование.
|
| id | journaliasakpiua-article-239871 |
| institution | System research and information technologies |
| keywords_txt_mv | keywords |
| language | English |
| last_indexed | 2025-07-17T10:27:26Z |
| publishDate | 2021 |
| publisher | The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" |
| record_format | ojs |
| resource_txt_mv | journaliasakpiua/57/c146a65448e0fd09021a5576ab9f0057.pdf |
| spelling | journaliasakpiua-article-2398712021-09-16T11:48:22Z Developing an algorithm for rapid assessment of living standards and quality of life of the population in the region Разработка алгоритма методики экспресс-оценки уровня и качества жизни населения региона Розроблення алгоритму методики експрес-оцінки рівня та якості життя населення регіону Lesnichaya, Marina Kolchina, Oksana Pahomov, Evgenii living standards variable reduction clustering discriminant analysis ranking уровень жизни населения тренды динамики редукция переменных кластеризация оценка качества классификации ранжирование рівень життя населення тренди динаміки редукція змінних кластеризація оцінювання якості класифікації ранжування The article presents the rapid assessment methodology that allows analysts to carry out qualitative monitoring of the living standards of the population using a wide range of methods for statistics processing (factor analysis, cluster analysis, discriminant analysis, method of combining indicators of different dimensions). This methodology is characterized by the high speed of mathematical calculations, availability to users with different skill levels and universal applicability to various study objects. The rapid assessment method is intended for screening the living standards of the population and activity quality of the territorial authorities, taking into account a different set of indicators. This article offers the author’s indicator system for assessing the living standards and quality of life of the population. The methodology algorithm describes flowcharts of the index method for combining statistical observations of different dimensions, which make it possible to automate the process of territory ranking. The study covers 12 urban districts and 43 municipal districts of Rostov Oblast. The methodology described in the article will help eliminate a subjective factor while monitoring, rationally distribute financial resources allocated annually by the authorities to support programs for socio-economic development of the territory, increase the economic efficiency and implementation speed of innovative projects that have a direct impact on the living standards and quality of life of the population. Предложена методика экспресс-оценки, позволяющая аналитикам проводить качественный мониторинг уровня жизни населения, используя широкий набор методов обработки статистики (факторный анализ, кластерный анализ, дискриминантный анализ, метод объединения разномерных показателей), отличающаяся высокой скоростью математических вычислений, доступностью пользователям с разным уровнем квалификации, универсальностью применения для различных объектов исследования. Метод экспресс-оценки предназначен для скрининга уровня жизни населения и качества деятельности органов власти территориальных образований, учитывающий различный набор показателей. Предложена авторская система показателей оценки уровня и качества жизни населения. В алгоритме методики описаны блок-схемы индексного метода объединения разномерных статистических наблюдений, позволяющие автоматизировать процесс ранжирования территорий. В качестве объекта исследования были выбраны 12 городских округов и 43 муниципальных района Ростовской области. Применение описанной методики позволит: исключить субъективный фактор при проведении мониторинга; рационально распределять финансовые ресурсы, ежегодно выделяемые органами власти на поддержку программ социально-экономического развития территории; повысить экономическую эффективность и скорость внедрения инновационных проектов, которые оказывают непосредственное влияние на уровень и качество жизни населения. Запропоновано методику експрес-оцінки, яка дозволить аналітикам проводити якісний моніторинг рівня життя населення, використовуючи широкий набір методів оброблення статистики (факторний аналіз, кластерний аналіз, дискримінантний аналіз, метод об’єднання різновимірних показників), що відрізняється високою швидкістю математичних обчислень, доступністю користувачам з різним рівнем кваліфікації, універсальністю застосування для різних об’єктів дослідження. Метод експрес-оцінки призначено для скринінгу рівня життя населення і якості діяльності органів влади територіальних утворень з урахуванням набору різних показників. Запропоновано авторську систему показників оцінювання рівня та якості життя населення. В алгоритмі методики описано блок-схеми індексного методу об’єднання різновимірних статистичних спостережень, що дозволяють автоматизувати процес ранжування територій. Як об’єкт дослідження обрано 12 міських і 43 муніципальних округів району Ростовської області. Застосування описаної методики дасть змогу: вилучити суб’єктивний фактор під час проведення моніторингу; раціонально розподіляти фінансові ресурси, які щорічно виділяються органами влади на підтримання програм соціально-економічного розвитку території; підвищити економічну ефективність і пришвидшити впровадження інноваційних проєктів, що безпосередньо впливають на рівень і якість життя населення. The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" 2021-09-14 Article Article application/pdf https://journal.iasa.kpi.ua/article/view/239871 10.20535/SRIT.2308-8893.2021.2.04 System research and information technologies; No. 2 (2021); 50-63 Системные исследования и информационные технологии; № 2 (2021); 50-63 Системні дослідження та інформаційні технології; № 2 (2021); 50-63 2308-8893 1681-6048 en https://journal.iasa.kpi.ua/article/view/239871/238268 |
| spellingShingle | рівень життя населення тренди динаміки редукція змінних кластеризація оцінювання якості класифікації ранжування Lesnichaya, Marina Kolchina, Oksana Pahomov, Evgenii Розроблення алгоритму методики експрес-оцінки рівня та якості життя населення регіону |
| title | Розроблення алгоритму методики експрес-оцінки рівня та якості життя населення регіону |
| title_alt | Developing an algorithm for rapid assessment of living standards and quality of life of the population in the region Разработка алгоритма методики экспресс-оценки уровня и качества жизни населения региона |
| title_full | Розроблення алгоритму методики експрес-оцінки рівня та якості життя населення регіону |
| title_fullStr | Розроблення алгоритму методики експрес-оцінки рівня та якості життя населення регіону |
| title_full_unstemmed | Розроблення алгоритму методики експрес-оцінки рівня та якості життя населення регіону |
| title_short | Розроблення алгоритму методики експрес-оцінки рівня та якості життя населення регіону |
| title_sort | розроблення алгоритму методики експрес-оцінки рівня та якості життя населення регіону |
| topic | рівень життя населення тренди динаміки редукція змінних кластеризація оцінювання якості класифікації ранжування |
| topic_facet | living standards variable reduction clustering discriminant analysis ranking уровень жизни населения тренды динамики редукция переменных кластеризация оценка качества классификации ранжирование рівень життя населення тренди динаміки редукція змінних кластеризація оцінювання якості класифікації ранжування |
| url | https://journal.iasa.kpi.ua/article/view/239871 |
| work_keys_str_mv | AT lesnichayamarina developinganalgorithmforrapidassessmentoflivingstandardsandqualityoflifeofthepopulationintheregion AT kolchinaoksana developinganalgorithmforrapidassessmentoflivingstandardsandqualityoflifeofthepopulationintheregion AT pahomovevgenii developinganalgorithmforrapidassessmentoflivingstandardsandqualityoflifeofthepopulationintheregion AT lesnichayamarina razrabotkaalgoritmametodikiékspressocenkiurovnâikačestvažizninaseleniâregiona AT kolchinaoksana razrabotkaalgoritmametodikiékspressocenkiurovnâikačestvažizninaseleniâregiona AT pahomovevgenii razrabotkaalgoritmametodikiékspressocenkiurovnâikačestvažizninaseleniâregiona AT lesnichayamarina rozroblennâalgoritmumetodikiekspresocínkirívnâtaâkostížittânaselennâregíonu AT kolchinaoksana rozroblennâalgoritmumetodikiekspresocínkirívnâtaâkostížittânaselennâregíonu AT pahomovevgenii rozroblennâalgoritmumetodikiekspresocínkirívnâtaâkostížittânaselennâregíonu |