Volume 7, No.

2 2019
ISSN 2303-5005

ANALYSIS OF MACROECONOMIC FACTORS EFFECT TO GROSS

DOMESTIC PRODUCT OF BOSNIA AND HERZEGOVINA USING
THE MULTIPLE LINEAR REGRESSION MODEL

Stanko Stanić, Željko V. Račić1

date of paper receipt: date of sending to review: date of review receipt:

27.11.2019. 29.11.2019. 06.12.2019.

Original Article doi: 10.2478/eoik-2019-0022 UDK: 330.101.541:332.155(497.6)

1
University of Banja Luka, Faculty of Economics, Bosnia and Herzegovina

ABSTRACT

This paper presents the application of the multiple regression analysis model in macroeconomic
research using the model of Bosnia and Herzegovina in the period from 2005 to 2018. The objective
of the research is to evaluate the effects of macroeconomic factors (independent variables) to gross
domestic product (dependent variable), and based on theoretical and methodological research.
Applying the Enter method, out of six independent variables, they are all included in the regression
model, whereas the sequence of inclusion in the model is the following: foreign direct investments,
Import, Export, Growth rate, unemployment and inflation. Numerous research indicate positive
connection between gross domestic product as the dependent variable and foreign direct
investments, Import, Export, Growth rate, unemployment and inflation, as independent variables.
Other factors negligibly explain the most important indicator of economic activities of a country.
Our assignment is to either confirm or reject the abovementioned statement.

Keywords:
Gross domestic product, multiple regression linear model, Enter method, determination coefficient.

JEL: C44

91
ECONOMICS
INTRODUCTION

Numerous factors have effect on economic growth and country development, and one of the most
important ones is gross domestic product (GDP). On the other hand, many factors also influence the
gross domestic product. The selection of these factors represents a subjective issue of the decision-
maker and the analyst (researcher). Besides these, there are also other factors important for the
functioning of economies: market openness, offer and demand for goods and services, unequal
income distribution, population, political situation etc. Natural factors have equally important
influence on economic growth: climate changes, diversification and sustainable development.
With the purpose of quantification of relation between economic growth and other macroeconomic
indicators, we have selected extended multiple linear regression model by applying Enter
method and regression procedures, as well as residual analysis. Enter method gradually includes
independent variables, and then one by one variable is removed from the model (in accordance
with the previously set criteria). The sequence of including independent variables in the model is
determined by the contribution of each specific independent in the explanation of the variability of
the dependent variable.

1. PREVIOUS RESEARCH

Economic growth includes the processes of implementing successful development strategies of a

country and determines the level of its wealth. The issue of relation between export, foreign direct
investments and economic growth has caused great attention and interest of both economists and
creators of economic politics worldwide.
Regardless of the fact whether a country is developed, in transition or under-developed, market
openness encourages economic growth (Etale, 2016).
When analysing the relations and possible effects of macroeconomic factors to GDP of the Western
Balkans countries, it is necessary to define the ”new growth model” (Estrin and Uvalic, 2013).
Taking into account unfavourable structure of foreign direct investments, export/import, instability
of political and social circumstances, (non)compliance of the legislation with EU regulations and
directives, as well as many other factors, one may not expect positive effects of analysed phenomena
to GDP in a short period of time.
The conducted panel analysis (Mahmoodi, M. and Mahmoodi E., 2016) indicates long-term
connection economic growth and export and foreign direct investments. Their results have indicated
a long-term two-way connection between foreign direct investments, export and economic growth.
The results of research on the effect of foreign direct investments to economic growth of countries
in transition (Nestorovic, 2015) based on regression analysis, show positive, but not statistically
significant correlation between foreign direct investments and economic growth in the countries
in transition.
To support this statement, we also point out the fact that the correlation between, for example,
foreign direct investments and economic growth in negative, since the dominant form of foreign
direct investments in the Western Balkan countries was acquisition. Acquisitions cannot be
considered as investments into real property, since the sale income can be used for expenditure
and investments, and in that way foreign direct investments cannot reflect on economic growth
(Mencinger, 2003).

2. APPLIED METHODOLOGY

Gross domestic product (dependent variable) has been analysed, and which is in function of foreign
direct investments, import, export, growth rate, unemployment and inflation, as independent
variables, in Bosnia and Herzegovina in the period (from 2005 to 2018). The statistical data have

92
 ECONOMICS
been retrieves from the Eurostat database. The analysis and model comparison is carried out by
using the programme IBM SPSS Statistics 23.

2.1 DEPENDENT VARIABLE

Gross domestic product. GDP – gross domestic product represents the total value of all the final
goods and services produced in a specific time period on a specific territory. GDP is stated in
monetary units, and includes only goods and services intended for further processing and
production. GDP indicates the measure of production activities and it is a generally accepted
indicator of the condition of a specific economy. 

2.2 INDEPENDENT VARIABLES

Foreign direct investments (FDI). Foreign investments include: foreign direct investments, joint
investments, portfolio investments, concessions and other forms, such as: greenfield and brownfield
investments, merging, absorptions etc.
GDP growth rate: Growth rate is the ratio between the GDP in the analysed year and the GDP in
previous year, in current prices. Increase of GDP is mostly influenced by the growth of foreign trade
balance, increase of export as opposed to import, and increase of consumption of the population.
Unemployment rate. It represents the participation of the unemployed in the total work power.
Total work power represents the sum of employed and unemployed people, who are actively seeking
work (aged between 17 to 74 years).
Export of goods and services. It includes transactions of goods and services from residents to
non-residents. Export has positive effect on GDP growth, and it indicates the activity of domestic
companies in the foreign market.
Import of goods and services. It includes transactions of goods and services from non-residents to
residents. Export has negative effect on creating GDP because it decreases the positive effects of
export and increases the dependency of one country on products from other countries.
Inflation. Inflation is measured using the index of consumer prices. The index of consumer prices
measures the changes of average level of prices of products and services per main groups of products
and services classified in accordance with their intention in consumption (personal consumption).

The selection of indicators has been carried out based on macroeconomic development indicators
that are available on the website of the European Statistical Office1.

1 http://ec.europa.eu/eurostat.

93
ECONOMICS
3. APPLICATION OF MULTIPLE LINEAR REGRESSION MODEL:
ANALYSIS OF BOSNIA AND HERZEGOVINA GDP

Data sources used for application of regression model

The analysis starts with specification of regression model elements which includes the selection of
exogenous variables and functional model forms.

Table 1. Macroeconimic indicators of Bosnia and Herzegovina economy

Nominal FDI Import Export Growth Unemployment Inflation

GDP rate
2005 17.650,00 552,3 12611,18 5.910,82 5,1 509.224,00 3,8
2006 20.057,00 864,8 13183,18 6.010,82 5,4 519.224,00 6,1
2007 22.548,00 2.599,80 13.811,18 6.110,82 5,9 520.014,00 1,5
2008 25.519,00 1.337,50 16.058,45 6.851,45 5,4 477.609,00 7,4
2009 24.799,00 351,9 12.634,02 6.201,96 -3 510.534,00 -0,4
2010 25.365,00 599,7 13.730,68 7.532,26 0,9 522.989,00 2,1
2011 26.231,00 698,6 15.640,25 8.403,44 1 536.728,00 3,7
2012 26.222,70 601,1 15.577,04 8.483,19 -0,8 550.255,00 2,1
2013 26.778,80 700 15.297,22 9.035,97 2,4 553.481,00 -0,1
2014 27.358,70 811 16.523,27 9.296,53 1,1 547.134,00 -0,9
2015 28.589,10 636,9 16.319,44 9.969,31 3,1 537.568,00 -1
2016 29.904,50 564 16.832,09 10.659,83 3,1 512.496,00 -1,1
2017 31.376,20 777,7 19.160,91 12.574,50 3,2 478.191,00 1,2
2018 32.954,00 800 20.139,38 13.565,16 3,2 437.783,00 1,4

Results of applying the multiple linear regression model

Analysis of multiple linear regression model is carried out with the IBM SPSS software package of
the latest version Statistics 23. From the exit results we pay attention to the key tables.

Table 2 Correlations, indicates the values of the Pearson correlation coefficient between all the
variables. Import is in the strongest correlation with Export (-0,944), following GDP with Export
(0,915), as well as GDP with Import (-0,895); and the weakest correlation is FDI with Import
(-0,067).

Table 2.

GDP FDI Import Export Growth Unem- Inflation

rate ployment
GDP 1,000 -,196 -, 895 ,915 -,257 -,332 -,501
FDI -,196 1,000 -,067 -,254 ,575 -,104 ,223
Import ,-895 -,067 1,000 -, 944 ,072 -,494 -,240
Pearson Export ,915 -,254 ,-944 1,000 -,064 -,403 -,454
Correlation Growth rate -,257 ,575 ,072 -,064 1,000 -,324 ,463
Unemploy- -,332 -,104 -,494 -,403 -,324 1,000 -,274
ment
Inflation -,501 ,223 -,240 -,454 ,463 -,274 1,000

94
 ECONOMICS
We used the Enter method, where the initial situation implies that all independent variables are
included in the model, without any restrictions. Then one by one variable is excluded from the
model according to a predefined criteria that is related to the value of the F to removal statistics. The
following table (Table 3) indicates that, according to the defined statistical criteria, all independent
variables were retained in the regression model.

Table 3. Variables Entered/Removeda

Variables
Model Variables Entered Method
Removed
Inflation, FDI, Import, Unemployment,
1 Enter
Growth Rate, Exportb
a. Dependent Variable: GDP
b. All requested variables entered.

We now move to Table 4, Summary Model, which provides key information about the validity
of the regression model. Table 4 presents two regression models: the first model containing an
independent variable and the second one with two independent variables.

Table 4. Model Summaryb

Model R R Square Adjusted R Std. Error Change Statistics Durbin-

Square of the R F df1 df2 Sig. F Watson
Estimate Square Change Change
Change
1 ,966a ,932 ,874 1459,59007 ,932 16,097 6 7 ,001 1,942
a. Predictors: (Constant), Inflation, FDI, Import, Unemployment, Growth rate, Export
b. Dependent Variable: GDP

Multiple correlation coefficient of the dependent variable (GDP) and independent variables (FDI,
Import, Export, Growth rate, Unemployment, Inflation) is 0,966, meaning that their connection is
very strong.
Determination coefficient is 0,932, meaning that 93,2 % of the variability of the dependent variable
(GDP) could be explained by the effect of the independent variables (FDI, Import, Export, Growth
rate, Unemployment, Inflation).
The value of the adjusted determination coefficient of 0,874 is close to the value of the regular
determination coefficient (0,932), due to favourable ratio between the number of variables included
in the model and the total number of observations.
Based on retrieved regression coefficients, our multiple linear regression model can be presented
as it follows:

where:
dependent variable (in our example - GDP),
independent variables (in our example – FDI, Import, Export, Growth rate, Unemployment,
Inflation),
Constant,
= unknown parameter with the independent variables (FDI, Import, Export, Growth rate,
Unemployment, Inflation),
error which reflects all the influences on the dependent variable (GDP) that do not originate from
the independent variables (FDI, Import, Export, Growth rate, Unemployment, Inflation)

95
ECONOMICS
Table 5 presents regular regression coefficients – under B and standardized regression coefficients
under.

Table 5. Coefficientsa

Stan-
Unstandardized Coef- dardized 95,0% Confidence Inter- Collinearity
Correlations
ficients Coeffi- val for B Statistics
Model cients t Sig.
B Std. Error Beta Lower Upper Zero- Partial Part Tol- VIF
Bound Bound order er-
ance
1 7476,930 11385,199 ,657 ,532 -19444,788 34398,648
(Constant)
FDI ,639 1,282 ,085 ,499 ,633 -2,391 3,670 -,196 ,185 ,049 ,329 3,040
Import 1,333 ,940 ,733 1,418 ,199 -,889 3,556 ,895 ,472 ,139 ,036 27,662
Export ,228 1,017 ,135 ,224 ,829 -2,177 2,633 ,915 ,084 ,022 ,027 37,572
Growth -468,346 242,380 -,290 -1,932 ,095 -1041,485 104,792 -,257 -,590 -,190 ,428 2,335
rate
Unempl. -,006 ,017 -,045 -,337 ,746 -,046 ,034 -,332 -,126 -,033 ,546 1,831
Inflation -251,522 321,310 -,161 -,783 ,459 -1011,299 508,255 -,501 -,284 -,077 ,229 4,372
a. Dependent Variable: GDP

In our example, and based on retrieved regression coefficients, our regression model can be
presented by the following equation:

Apart from the written model, it can be further concluded that out of all the analyzed independent
variables, Import has major effect on GDP, followed by FDI and Export.

CONCLUSIONS

Economic theory and practice do not know any unique set of indicators that affect GDP. The very
complexity of this phenomenon relativises any attempt to establish an unchanged list of indicators.
Therefore, in selection of these indicators is based on subjectivity of the authors. The actual number
of indicators remains to be seen in the specific analysis, taking into account their interdependence
and the occurrence of duplicity, since these two facts directly influence the selection of the optimal
number of indicators.
With the purpose of quantification of relation between macroeconomic factors and GDP, we have
used multiple linear regression model. The analysis showed that GDP with as much as 93.2%
can be explained by the influence of independent variables: FDI, Import, Export, Growth rate,
Unemployment, Inflation. Other factors negligibly (6.8%) explain the most important indicator of
economic activities of a country.
Out of all the analysed independent variables, Import has major effect on GDP, followed by FDI
and Export.
From the abovementioned we can confirm the set hypothesis of this paper that the selection of
predictors defined in the multiple linear regression model is significant, in terms of Bosnia and
Herzegovina GDP development over the observed period.
The research results may be useful to developers of economic policies for bringing the right
decisions in solving external shocks and crisis, in order to focus their activities more precisely in
the direction of economic growth and overall progress of Bosnia and Herzegovina economy.

96
 ECONOMICS
REFERENCES

[1] Buterin V., Škare, M., Buterin, D. (2017). Macroeconomic model of institutional reforms’ influence
on economic growth of the new EU members and the Republic of Croatia. Economic Research-
Ekonomska Istraživanja, 30:1, p. 1572-1593, DOI: 10.1080/1331677X.2017.1355260.
[2] Estrin, S., Uvalic, M. (2013). Foreign direct investment into transition economies: Are the Balkans
different? London School of Economics and political Science, England.
[3] Etale, E. L., Etale, M., Lyndon, M. (2016). The Relationship between Exports, Foreign Direct
Investment and Economic Growth in Malaysia, International Journal of Business Management and
Economic Research (IJBMER), 7(2), p. 572–578. ISSN: 2229-6247.
[4] Mahmoodi, M., Mahmoodi, E. (2016). Foreign direct investment, exports and economic
growth: evidence from two panels of developing countries. Economic Research, 29:1, 938-949. DOI:
10.1080/1331677X.2016.1164922.
[5] Mano-Bakalinov, V. (2016). Trade liberalisation and economic growth in Macedonia. South East
European Journal of Economics and Business. Volume 11 (2), 48-60. DOI: 10.1515/jeb-2016-0010.
[6] Mencinger, J. (2003). Does Foreing Investment Always Enhance Economic Growth? International
Review for Social Sciences, 56(4):491–508; p. 534-547.
[7] Nestorović, O. (2015). Da li strane direktne investicije doprinose privrednom rastu zemalja u
tranziciji?, Ekonomske teme 53(2), 273–282.UDK 339.727.22:338.1.
[8]Pere, E. (2015). The impact of good governance in the economic development of Western Balkan
countries. European Journal of Government and Economics, Volume 4: 25-45, Retrieved from
http://www.ejge.org/index.php/ejge/article/view/59/52.
[9] World Bank, (2018). World Development Indicators (WDI).

Retrieved from:
http://databank.worldbank.org/data/reports.aspx?source=worlddevelopment- indicators.

97