Use of GMDH for Investigation of Impact of Non-income Components on HDI
Impact of macroeconomic indices on the development level of countries is analysed in the article. UN data about human development for 2011 are used. Countries are divided into groups according to UN methodology. The factors influencing on the development level for every group were chosen and analyse...
Збережено в:
| Дата: | 2012 |
|---|---|
| Автори: | , |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Міжнародний науково-навчальний центр інформаційних технологій і систем НАН та МОН України
2012
|
| Назва видання: | Індуктивне моделювання складних систем |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/45955 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Цитувати: | Use of GMDH for Investigation of Impact of Non-income Components on HDI / E. Savchenko, O. Tutova // Індуктивне моделювання складних систем: Зб. наук. пр. — К.: МННЦ ІТС НАН та МОН України, 2012. — Вип. 4. — С. 28-37. — Бібліогр.: 8 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-45955 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-459552013-06-22T03:15:15Z Use of GMDH for Investigation of Impact of Non-income Components on HDI Savchenko, E. Tutova, O. Impact of macroeconomic indices on the development level of countries is analysed in the article. UN data about human development for 2011 are used. Countries are divided into groups according to UN methodology. The factors influencing on the development level for every group were chosen and analysed. Combinatorial GMDH algorithm was used to build models describing dependence of human development index on macroeconomic factors. У статті аналізується вплив макроекономічних показників на рівень людського розвитку держав світу. Взято дані ООН про розвиток країн за 2011 рік. Країни поділені на групи відповідно до класифікації ООН. Виділено та проаналізовано показники, які впливають на рівень розвитку кожної з груп. Для побудови моделей, що описують залежність індексу розвитку людського потенціалу від макроекономічних показників, використаний комбінаторний алгоритм МГУА. В статье анализируется влияние макроэкономических показателей на уровень человеческого развития государств мира. Взяты данные ООН о развитии стран за 2011 год. Страны поделены на группы в соответствии с исследованиями ООН. Выделены и проанализированы показатели, которые влияют на уровень развития каждой из групп. Для построения моделей, описывающих зависимости индекса развития человеческого потенциала от макроэкономических показателей, использован комбинаторный алгоритм МГУА 2012 Article Use of GMDH for Investigation of Impact of Non-income Components on HDI / E. Savchenko, O. Tutova // Індуктивне моделювання складних систем: Зб. наук. пр. — К.: МННЦ ІТС НАН та МОН України, 2012. — Вип. 4. — С. 28-37. — Бібліогр.: 8 назв. — англ. XXXX-0044 http://dspace.nbuv.gov.ua/handle/123456789/45955 681.513; 004.942 en Індуктивне моделювання складних систем Міжнародний науково-навчальний центр інформаційних технологій і систем НАН та МОН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| description |
Impact of macroeconomic indices on the development level of countries is analysed in the article. UN data about human development for 2011 are used. Countries are divided into groups according to UN methodology. The factors influencing on the development level for every group were chosen and analysed. Combinatorial GMDH algorithm was used to build models describing dependence of human development index on macroeconomic factors. |
| format |
Article |
| author |
Savchenko, E. Tutova, O. |
| spellingShingle |
Savchenko, E. Tutova, O. Use of GMDH for Investigation of Impact of Non-income Components on HDI Індуктивне моделювання складних систем |
| author_facet |
Savchenko, E. Tutova, O. |
| author_sort |
Savchenko, E. |
| title |
Use of GMDH for Investigation of Impact of Non-income Components on HDI |
| title_short |
Use of GMDH for Investigation of Impact of Non-income Components on HDI |
| title_full |
Use of GMDH for Investigation of Impact of Non-income Components on HDI |
| title_fullStr |
Use of GMDH for Investigation of Impact of Non-income Components on HDI |
| title_full_unstemmed |
Use of GMDH for Investigation of Impact of Non-income Components on HDI |
| title_sort |
use of gmdh for investigation of impact of non-income components on hdi |
| publisher |
Міжнародний науково-навчальний центр інформаційних технологій і систем НАН та МОН України |
| publishDate |
2012 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/45955 |
| citation_txt |
Use of GMDH for Investigation of Impact of Non-income Components on HDI / E. Savchenko, O. Tutova // Індуктивне моделювання складних систем: Зб. наук. пр. — К.: МННЦ ІТС НАН та МОН України, 2012. — Вип. 4. — С. 28-37. — Бібліогр.: 8 назв. — англ. |
| series |
Індуктивне моделювання складних систем |
| work_keys_str_mv |
AT savchenkoe useofgmdhforinvestigationofimpactofnonincomecomponentsonhdi AT tutovao useofgmdhforinvestigationofimpactofnonincomecomponentsonhdi |
| first_indexed |
2025-07-04T05:00:13Z |
| last_indexed |
2025-07-04T05:00:13Z |
| _version_ |
1836691194596294656 |
| fulltext |
Use of GMDH for Investigation of Impact
Індуктивне моделювання складних систем, випуск 4, 2012 28
УДК 681.513; 004.942
USE OF GMDH FOR INVESTIGATION OF IMPACT OF
NON-INCOME COMPONENTS ON HDI
Savchenko E., Tutova O.
International Research and Educational Center of Information Technologies and Systems of
NAS of Ukraine, pr. Academika Glushkova, 40, Kyiv, 03680, Ukraine
savchenko@irtc.org.ua, sir_ludovick@yahoo.com
У статті аналізується вплив макроекономічних показників на рівень людського розвитку держав світу. Взято
дані ООН про розвиток країн за 2011 рік. Країни поділені на групи відповідно до класифікації ООН. Виділено
та проаналізовано показники, які впливають на рівень розвитку кожної з груп. Для побудови моделей, що
описують залежність індексу розвитку людського потенціалу від макроекономічних показників, використаний
комбінаторний алгоритм МГУА.
Ключові слова: комбінаторний алгоритм МГУА, індекс людського розвитку, макроекономічні показники
Impact of macroeconomic indices on the development level of countries is analysed in the article. UN data about human
development for 2011 are used. Countries are divided into groups according to UN methodology. The factors
influencing on the development level for every group were chosen and analysed. Combinatorial GMDH algorithm was
used to build models describing dependence of human development index on macroeconomic factors.
Keywords: combinatorial GMDH algorithm, human development index, macroeconomic data
В статье анализируется влияние макроэкономических показателей на уровень человеческого развития
государств мира. Взяты данные ООН о развитии стран за 2011 год. Страны поделены на группы в соответствии
с исследованиями ООН. Выделены и проанализированы показатели, которые влияют на уровень развития
каждой из групп. Для построения моделей, описывающих зависимости индекса развития человеческого
потенциала от макроэкономических показателей, использован комбинаторный алгоритм МГУА.
Ключевые слова: анализ, комбинаторный алгоритм МГУА, индекс развития человеческого,
макроэкономические показатели
1. Introduction
United Nations Development Program (UNDP) is the United Nation's (UN)
global development network, advocating for change and connecting countries to
knowledge, experience and resources to help people build a better life [1]. UNDP
works in four main areas: poverty reduction; democratic governance; crisis
prevention and recovery; environment and sustainable development.
UN specialists annually prepare reports about current development of the
countries, where all countries are ranked by human development index [2].
The goal of this work is to analyze what factors affect human development level
by building dependencies by the group method of data handling (GMDH) and to
analyze received result from the point of view of economic situation. Such models
may be used by specialists for current economic situation analysis of the country in
order to find possibilities for improvement of level of country development,
particularly Ukraine. Data for 2011 were used for analysis.
Besides, it was shown how GMDH can be used for revealing of dependencies in
social and economic data and their analysis. Models are developed by Combinatorial
GMDH algorithm [3, 4]. Combinatorial GMDH algorithm is a method of structural-
parametric identification when structure and model parameters are selected in the
process of model building unlike regression analysis that uses a fixed model
Savchenko E., Tutova O.
Індуктивне моделювання складних систем, випуск 4, 2012 29
structure. Inductive GMDH algorithms give possibility to find interrelations in data
automatically, and to select optimal structure of model.
Models for forecasting of human development index of Ukraine were developed
in the works [5-8]. This forecast enables to see the change of index in the future and
can be used for analysis by specialists. GMDH is the method which enables building
of long-term forecast. The models built by this method using external bias criterion
are more noise immunity than ones built by other methods.
The task of the present work is to find out what factors are the most influential for
human development in countries with different development levels. Is income the
main factor for human well being? There are assumptions that people in highly
developed countries benefit from reliable social system and comprehensive public
health system. Their human development potential lies in the field of non-income
stimulus. While population of developing countries suffers from poverty and often
have no resources to satisfy their basic needs. So the income factor is the most
important for them for achieving minimum standards of living. Therefore, the
purpose of this work is to verify such assumptions as well.
2. Human development index as factor of development level of countries
Human development can be defined as enabling people to develop their full potential
and lead productive, creative lives in accordance with their needs and interests.
However, it took a long time before mankind accepted the rather simple truth that the
goal of development is to enhance everyone’s abilities and freedoms. Over time there
has been a better understanding of the social consequences of economic development,
of the increasing inequality between rich and poor countries that accompanied
globalization and above all an acknowledgement by governments and the public at
large that not only is human development achievable, but that it has practical
meaning for social and economic progress and the overall prosperity of nations and
states. Therefore, the dynamics of human development analysis have become an issue
of research for such leading world organizations as the UN, World Bank and other
international organizations.
The Human Development Report is an independent publication commissioned by
the United Nations Development Programme. Its editorial autonomy is guaranteed by
a special resolution of the General Assembly which recognizes the Human
Development Report as “an independent intellectual exercise” and “an important tool
for raising awareness about human development around the world" [1].
Twenty years ago, the Human Development Index (HDI) was proposed as an
alternative to conventional assessments of development based on measures of per
capita income. It complements income with health and education indicators. Human
Development Index classifications are relative – based on quartiles of HDI
distribution across countries and denoted very high, high, medium and low HDI.
Because there are 187 countries, the four groups do not have the same number of
Use of GMDH for Investigation of Impact
Індуктивне моделювання складних систем, випуск 4, 2012 30
countries: the very high, high and medium HDI groups have 47 countries each, and
the low HDI group has 46 countries.
The components of HDI are life expectancy at birth, mean years of schooling,
expected years of schooling, gross national income (GNI) per capita, GNI per capita
rank minus HDI rank, and non-income HDI value.
Table 1 presents a fragment of data sample describing Human Development Index
and its components for the group of countries with very high human development in
2011.
Table 1. Fragment of input data sample for the countries with very high human
development in 2011
№ Country 1X 2X 3X 4X 5X 6X 1Y
1 Norway 81,1 12,6 17,3 47 557 6 0,975 0,943
2 Australia 81,9 12,0 18,0 34 431 16 0,979 0,929
3 Netherlands 80,7 11,6 16,8 36 402 9 0,944 0,910
4 United States 78,5 12,4 16,0 43 017 6 0,931 0,910
5 New Zealand 80,7 12,5 18,0 23 737 30 0,978 0,908
6 Canada 81,0 12,1 16,0 35 166 10 0,944 0,908
7 Ireland 80,6 11,6 18,0 29 322 19 0,959 0,908
8 Liechtenstein 79,6 10,3 14,7 83 717 -6 0,877 0,905
9 Germany 80,4 12,2 15,9 34 854 8 0,940 0,905
10 Sweden 81,4 11,7 15,7 35 837 4 0,936 0,904
11 Switzerland 82,3 11,0 15,6 39 924 0 0,926 0,903
12 Japan 83,4 11,6 15,1 32 295 11 0,940 0,901
13 Hong Kong, China (SAR) 82,8 10,0 15,7 44 805 -4 0,910 0,898
14 Iceland 81,8 10,4 18,0 29 354 11 0,943 0,898
15 Korea (Republic of) 80,6 11,6 16,9 28 230 12 0,945 0,897
16 Denmark 78,8 11,4 16,9 34 347 3 0,926 0,895
17 Israel 81,6 11,9 15,5 25 849 14 0,939 0,888
18 Belgium 80,0 10,9 16,1 33 357 2 0,914 0,886
19 Austria 80,9 10,8 15,3 35 719 -4 0,908 0,885
20 France 81,5 10,6 16,1 30 462 4 0,919 0,884
1X – life expectancy at birth (years);
2X - mean years of schooling (years);
3X - expected years of schooling (years);
4X - gross National Income (GNI) per capita (Constant 2005 PPP$);
5X - GNI per capita rank minus HDI rank;
Savchenko E., Tutova O.
Індуктивне моделювання складних систем, випуск 4, 2012 31
6X - nonincome HDI;
1Y - Human Development Index (HDI).
Thus, samples for the all four groups of countries are obtained.
The values of all these components should be scaled to build an adequate model
of dependences of HDI on macroeconomic data. For this purpose each factor’s data
are divided by its maximum value.
2. Dependences of HDI construction by Macroeconomic Data
Combinatorial GMDH algorithm was chosen for description of dependences of HDI
on its components. It is a method of automatic model building based on data
observation [3, 4]. This method is grounded on the principles of induction, which is
the transition from specific to general. Unlike regression analysis which uses a fixed
model structure, combinatorial GMDH algorithm is a method of structural-parametric
identification.
For every group of countries models describing dependency of HDI on other
factors were built by GMDH algorithm.
Therefore, for the HDI 1, which characterizes a group of countries with very high
level of development, the following model was developed.
65431 9087,00312,00815,00035,00345,0 XXXXY +−+−= .
Value of criteria: AR = 0,68649E-04; BS = 0,6439E-05,
when AR is regularity external criterion, BS is bias external criterion [4].
Dependence of real and modelled HDI values on macroeconomic data for a group
of countries with very high level of development is shown in the Fig. 1.
0.790
0.810
0.830
0.850
0.870
0.890
0.910
0.930
0.950
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46
№
HDI
Hdi real Hdi model
Fig. 1. Comparison of Real HDI 1 and Modeled HDI 1
Use of GMDH for Investigation of Impact
Індуктивне моделювання складних систем, випуск 4, 2012 32
Similar model was built for the second group of countries with high development
level:
63212 0242,01331,02242,06998,00469,0 XXXXY ++++−=
Value of criteria: AR = 0,000059416; BS = 0,14473E-04.
Fig. 2 shows obtained dependence of real and modeled values for 47 countries
from the group with high development level.
0.430
0.480
0.530
0.580
0.630
0.680
0.730
0.780
0.830
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47
№
HDI
Hdi real Hdi model
Fig. 2. Comparison of Real HDI 2 and Modeled HDI 2
Expected years of schooling are a number of years of schooling that a child of
school entrance age can expect to receive if prevailing patterns of age-specific
enrolment rates persist throughout the child’s life. Mean years of schooling is an
average number of years of education received by people ages 25 and older,
converted from education attainment levels using official durations of each level.
The following model was developed for the third group of countries with middle
development level
6543213 0319,00632,01237,02175,04312,03343,00869,0 XXXXXXY ++++++−=
Value of criteria: AR = 0,000090547; BS = 0,548818E-04.
Fig. 3 shows dependence of real and modeled values for 47 countries from the
group with middle development level.
Savchenko E., Tutova O.
Індуктивне моделювання складних систем, випуск 4, 2012 33
0.300
0.400
0.500
0.600
0.700
0.800
0.900
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46
№
HDI
Hdi real Hdi model
Fig. 3. Comparison of Real HDI 3 and Modeled HDI 3
Countries with low development level are included in the fourth group. The
following model is developed for them:
4324 2689,03747,07612,02688,0 XXXY +++−=
Value of criteria: AR = 0.000189805; BS = 0.15863653E-03,
Dependence of real and modeled HDI values for the 47 countries from the fourth
group is shown on the Fig. 4.
0.300
0.400
0.500
0.600
0.700
0.800
0.900
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46
№
HDI
Hdi real Hdi model
Fig. 4. Comparison of Real HDI 4 and Modeled HDI 4
Use of GMDH for Investigation of Impact
Індуктивне моделювання складних систем, випуск 4, 2012 34
Life expectancy at birth is a number of years a newborn infant could expect to
live if prevailing patterns of age-specific mortality rates at the time of birth stay the
same throughout the infant’s life.
GNI per capita rank minus HDI rank is the difference in rankings by GNI per
capita and by the HDI.
Non-income HDI is the value of the HDI computed from the life expectancy and
education indicators only.
As we can see from Fig. 1 – 4, models built with GMDH algorithm describe
HDIs’ dependences on their components very preciously in groups of countries with
very high and high levels of development. To build an efficient model for the
countries with medium and low level of development we should extend our sample. It
is calculated just with health and education – and thus without the income
component. In other words, the value-added part of HDI is precisely in its non-
income components.
Since factors influencing HDI were analyzed, data sample was extended with new
factors which may affect development level of countries from these groups.
7X – share of multidimensional poor with deprivations in environmental services:
clean water;
8X – share of multidimensional poor with deprivations in environmental services:
improved sanitation;
9X – share of multidimensional poor with deprivations in environmental services:
modern fuels;
10X – ecological footprint (hectares per capita);
11X – greenhouse gas emissions per capita;
12X – urban pollution.
13X – satisfaction with air quality (% satisfied);
14X – satisfaction with water quality (% satisfied);
15X – urban (% of total).
It should be noticed that data presented in UN report contain quite a lot of gaps,
so only those countries with full data were chosen.
Thus, the following model of HDI was obtained for 24 countries with medium
level of development:
12119
86531
*
3
018,00105,00275,0
029,002446,0099,0059,006186,1054906,0
XXX
XXXXXY
−−−
++−++−=
Value of criteria: AR = 0,00009; BS = 0,5488E-04.
Savchenko E., Tutova O.
Індуктивне моделювання складних систем, випуск 4, 2012 35
0.7
0.75
0.8
0.85
0.9
0.95
1
1.05
1 3 5 7 9 11 13 15 17 19 21 23 25 27
№
HDI
HDI real HDI model
Fig. 5. Compari son of Real HDI 3 and Modeled HDI 3 Based on Extended
Data Sample
The following model of HDI was obtained for 27 countries with low level of
development
151412
115432
*
4
0081,00167,00259,0
013,00187,02245,04023,07266,02127,0
XXX
XXXXXY
++−
−−−+++−=
Value of criteria: AR = 0,0002; BS = 0,9425.
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1 3 5 7 9 11 13 15 17 19 21 23 25 27
№
HDI
HDI real HDI model
Fig. 6. Comparison of Real HDI 4 and Modeled HDI 4 Based on Extended Data
Sample
Use of GMDH for Investigation of Impact
Індуктивне моделювання складних систем, випуск 4, 2012 36
3. Obtained Dependences Analysis for Groups of Countries with Different
Levels of Development
As we can see non-income HDI is the most influential factor for the first group of
states. So population of countries with very high development level seeks for
potential in further achievements predominantly not in the material well-being. They
are more interested in such non-income components of HDI as health and education.
Problems of healthy style of life, relationships, civil liability are the most important
for them. They consider economic challenges rather from the point of view of
possible threats of illegal migration, negative influence of global economic and
political processes, interferences of international organizations into the internal
affairs of their countries.
Rather high level of income allows people from the second group to enjoy their
well-being. From the other side, inequality in access to medical and educational
services, gap between rural and urban population often prevent them from living
long. As high value of factor of life expectancy at birth for HDI 2 shows, that the
same factors negatively influence on the human development level and life span as
well.
Refined model of HDI 3 reveals even stronger dependence of their human
development level on life expectancy at birth than for the countries from the second
group. Taking into consideration low level of income, challenges faced by people
from the third group become more menacing. Also the combination of high infant
mortality and deaths in young adulthood from accidents, wars, and childbirth, may
lower the overall life expectancy in this group.
Mean years of schooling is the most important factor for human development in
the fourth group. Refined model emphases on importance of factors connected with
schooling. Education is a powerful asset in overcoming poverty. Adult literacy is still
unachieved goal for a lot of people in the poorest countries. Going to school means to
provide a child with minimum standards of living.
People are the real wealth of nations. Impressed by the rise and fall of national
incomes (as measured by GDP), we tend to equate human welfare with material
wealth. The importance of GDP growth and economic stability should not be
understated: both are fundamental to sustained human progress, as is clear in the
many countries that suffer from their absence. But the ultimate yardstick for
measuring progress is people’s quality of life.
Therefore, from the stated above it can be concluded that for countries from all
four groups income is not the most influential factor for human development. The
GNI per capita turned out not to be the decisive factor for any group of countries
even with low and middle level of development. Obtained dependences show, that
healthy and education are the most necessary assets for human development potential
for people from all over the world.
Savchenko E., Tutova O.
Індуктивне моделювання складних систем, випуск 4, 2012 37
Therefore the human development includes not only economic development, but
also the basic capabilities for living a long and healthy life, being educated, having a
decent standard of living, enjoying political and civil freedoms to participate in the
life of one’s community, especially the enforcement of human rights, while also
seeking to preserve a healthy environment.
Summary
Macroeconomic data were analysed in the article. Use of GMDH algorithm gave
possibility to build an adequate model of dependences of HDIs on their components
according to the levels of human development in the four groups of countries. The
most influential factors were revealed for each group. Non-income part of HDI turned
out to be the most important for human development in the whole world.
References
1. http://www.undp.org.ua
2. http://hdr.undp.org/en/reports
3. Ivakhnenko А.Г. Inductive methods of the difficult systems self-organization.
– Kyiv : Nauk. dumka, 1982. – 296 p. (In Russian).
4. Ivakhnenko A. G., Stepashko V.S.: Pomekhoustoichivost modelirovania
(Noise-immunity of modeling). Kiev: Naukova Dumka, 216 p, 1985 (in Russian).
5. Ивахненко А.Г., Савченко Е.А., Ивахненко Г.А., Надирадзе А.Б.,
Тоценко В.Г. Самоорганизация моделей социального развития стран по данным
мониторинга // Нейрокомпьютеры: разработка и применение, №2, 2003, C. 39-
47.
6. Kondyrina A., Savchenko Ie., Dyrektorenko O. Forecasting of Ukraine
Human Development Index by Combinatorial GMDH Algorithm // Proceedings of
International Workshop of Inductive Modelling. Prague, 22-26 September. – 2007.–
P. 46 – 50.
7. Савченко Є.А. Директоренко О.В. Прогнозування соціально-
економічних процесів на основі МГУА // І Міжнародна науково-технічна
конференція „Обчислювальний інтелект (результати, проблеми, перспективи)”,
10–13 травня 2011. – Черкаси. – 2011. – С.234.
8. Савченко Є.А., Кондиріна А.Г., Директоренко О.В. Застосування
алгоритму МГУА для прогнозування індексу розвитку людського потенціалу
України // Індуктивне моделювання складних систем. Збірник наукових праць.
– К.: МННЦ ІТтаС НАН та МОН України, 2011. – Вип.2. – С. 175-182.
|