INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 06,2025
Journal:
https://www.academicpublishers.org/journals/index.php/ijai
page 2200
THE IMPORTANCE OF THE TOURISM INDUSTRY IN THE GROSS
DOMESTIC PRODUCT
Khoshimjon Khamidovich Nurullaev
Senior Lecturer at the Department of Economics and Tourism,
Oriental University
hoshimnurullaev@gmail. com
Tolagan Batirovich Sadikov
Senior Lecturer at the Department of Economics and Tourism,
Oriental University
Sadikovtolagan@gmail. com
Abstract:
Due to the complex nature of the relationship between the tourism industry and GDP,
economists have not reached a unanimous consensus on how these variables influence each
other. Therefore, this study was conducted to examine the impact of the tourism sector on GDP.
The research focuses on Turkey, Greece, Italy, Spain, Egypt, and Cyprus over the period 2003–
2022. To analyze these relationships, a regression analysis was carried out using the OLS
method in STATA-14. The results indicate that all regressors are statistically significant,
demonstrating that the development of the tourism sector contributes to GDP growth.
Keywords:
OLS method, GDP, tourism, Breusch-Pagan test, Breusch-Godfrey test, correlation,
regression
Introduction
Currently, the tourism sector has emerged as one of the fastest-growing industries
worldwide. Following oil refining and extraction and automobile manufacturing, tourism ranks
as the third-largest sector globally. As Aleksandrova (2002) states, tourism encompasses “the
aggregate of relationships and phenomena arising when people travel and stay in places
different from their permanent residence and work” [1]. Bystrov and Vorontsova (2007) define
the tourism services market as “an economic phenomenon that integrates demand and supply of
specific tourism goods and services in a given place and time” [2]. Today, people’s movements,
attitudes, and ways of seeking enjoyment in life are transforming. Individuals increasingly aim
to spend leisure time meaningfully, explore new cultures and traditions, and improve their well-
being—services that the tourism sector readily provides. Over recent years, this drive for
discovery has intensified, further accelerating the growth of tourism. In some countries, the
tourism industry is growing steadily, with annual growth rates reaching 8–10%, underscoring
its crucial role in national economies.
Methodology
In order to achieve the aforementioned research objectives, it is essential to construct an
appropriate econometric model and to compile a suitable dataset. Building such a model
necessitates the use of socio-economic and statistical indicators highlighted in previous
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 06,2025
Journal:
https://www.academicpublishers.org/journals/index.php/ijai
page 2201
literature. The selection of model specifications and variables, as well as their sources and
theoretical justifications, are presented below.
Several studies emphasize the importance of including variables such as tourism
revenues, the number of international tourists, and macroeconomic indicators like GDP,
exchange rates, and inflation in tourism-related econometric models (Balaguer & Cantavella-
Jordá, 2002; Brida et al., 2016). Moreover, panel data analysis and regression techniques such
as OLS, fixed-effects, or random-effects models are widely applied to examine these
relationships across countries and time periods (Chen & Chiou-Wei, 2009; Gunduz & Hatemi-J,
2005).
For this study, annual data covering the period from 2003 to 2022 were collected from
official sources such as the World Bank, UNWTO, national statistical institutes, and central
banks of the countries analyzed. The model specification was developed considering theoretical
insights and empirical findings from prior research to ensure robust and meaningful results.
=+ +
(1.1)
In this study, the main indicator representing Gross Domestic Product (GDP) is derived
from the World Bank’s data repository, reflecting each country’s annual GDP figures over the
selected period. Descriptive statistics for this variable are presented after introducing the model
variables. In the STATA commands, the variable name is designated as “GDP” to ensure
consistency and ease of interpretation when working with commands and reporting results.
GDP serves as the primary dependent variable in the econometric model.
As the key explanatory variable potentially influencing GDP, the tourism indicator
(“tourism”) was selected. Data for this variable were also obtained from the World Bank
database, representing the volume of tourism activity in each country. Within STATA, this
variable is coded as “tourism” (“number”), facilitating straightforward usage in the statistical
software.
Descriptive statistics for the variables are presented in Table 1. According to the table, the
average number of tourists across the selected countries during the observed period amounts to
approximately 37,482,519 annually.
Table 1. Descriptive statistics of variables
Variable
Obs
Mean
Std. Dev.
Min
Max
number
100
41643916
37482519
2370000
1.262e+08
GDP
120
7.365e+11
6.581e+11
1.719e+10
1.992e+12
Greater emphasis is placed on examining the descriptive statistics of the key variables. It
is crucial to assess the distribution of the datasets and to determine the extent to which they
approximate a normal distribution. To achieve this, histograms of the variables are generated in
STATA and compared against the theoretical normal distribution curve. This approach provides
valuable insights into the data’s underlying patterns and the suitability of the econometric
methods applied.
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 06,2025
Journal:
https://www.academicpublishers.org/journals/index.php/ijai
page 2202
0
5.
0e
-1
3
1.
0e
-1
2
1.
5e
-1
2
D
en
si
ty
0
5.000e+11
1.000e+12
1.500e+12
2.000e+12
yaim
0
1.
0e
-0
8
2.
0e
-0
8
3.
0e
-0
8
D
en
si
ty
0
50000000
1.000e+08
1.500e+08
soni
Diagram 1. Distribution of GDP indicator
Diagram 2 Distribution of tourism volume indicator
Figure 1 illustrates the distribution of the GDP variable, plotted alongside a normal
distribution curve constructed based on the sample mean and standard deviation. From the
figure, it is evident that the GDP data do not follow a normal distribution.
Figure 2 presents the frequency distribution for the number of tourists. Similarly, the
diagram shows that the tourism variable does not conform to a normal distribution pattern.
To gain preliminary insights into the direction and strength of the relationship between
the two key economic indicators, Figure 3 was generated. This figure takes the form of a
scatterplot depicting the relationship between GDP and the number of tourists. The scatterplot
indicates a positive association between GDP and tourism numbers, as evidenced by the upward
trend in the distribution of data points.
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 06,2025
Journal:
https://www.academicpublishers.org/journals/index.php/ijai
page 2203
0
5.
00
0e
+1
1
1.
00
0e
+1
2
1.
50
0e
+1
2
2.
00
0e
+1
2
ya
im
0
50000000
1.000e+08
1.500e+08
soni
Diagram 3 Relationship between GDP and tourism volume
However, it is important to recognize that the observed relationship might also be
influenced by other factors. Therefore, a deeper analysis of the connection between these two
variables is warranted. To achieve this, calculating the correlation coefficients between the
variables is considered appropriate.
Table 2
presents the correlation matrix.
An examination of the table reveals that there is indeed a positive economic relationship
between the two indicators. The correlation coefficient, calculated at
0.812
, confirms the pattern
observed in the scatterplot. Based on the correlation coefficients, we can conclude that there is a
strong economic relationship between GDP and tourism figures.
Table 2. Matrix of correlation coefficients
Variables (1)
(2)
(1) GDP
1.000
(2)
number
0.812
1.000
Taking these aspects into consideration, the primary objective of this study is to
determine the interrelationship between the economic indicators under investigation through a
regression model. To achieve this goal, calculations were performed using the Ordinary Least
Squares (OLS) method within the STATA-14 software environment. The results and their
subsequent discussion are presented in the following section.
Results Analysis
When discussing the outcomes of the conducted regression analysis, it is first essential
to examine the initial regression results. These preliminary findings are presented in Table 3
.
The table indicates that both individual and overall statistical significance levels of the
estimated coefficients are below 0.05, implying that the variables are statistically significant.
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 06,2025
Journal:
https://www.academicpublishers.org/journals/index.php/ijai
page 2204
According to the regression results, the number of tourists exerts a positive influence on
GDP. The analysis demonstrates that tourism has a substantial and statistically significant effect
on GDP growth.
Table 3. Initial regression results
GDP
Coef.
St.Err.
t-
value
p-
value
[95%
Conf
Interval]
Sig
number
14412.26
6
1046.163 13.78 0
12336.19 16488.34
2
***
Constant
1.638e+1
1
5.848e+1
0
2.80
.006
4.774e+1
0
2.799e+1
1
***
Mean dependent var
763977662686.
483
SD dependent var
665217206952.
391
R-squared
0.659
Number of obs
100
F-test
189.787
Prob > F
0.000
Akaike crit. (AIC)
5623.733
Bayesian crit. (BIC)
5628.944
*** p<.01, ** p<.05, * p<.1
The presence of heteroskedasticity issues is highly significant when assessing the
quality of an econometric model. Therefore, the Breusch-Pagan tests for
heteroskedasticity were conducted, with the results presented in Table 4
.
Table 4. Brusch-Peagan heteroscedasticity test
Breusch-Pagan / Cook-Weisberg test for heteroskedasticity
Ho: Constant variance
Variables: fitted values of yaim
chi2(1)
= 28.05
Prob > chi2 = 0.0000
The Breusch-Pagan test indicates that the null hypothesis of homoskedastic error terms
is rejected. Based on the outcomes of these tests, it can be concluded that the model suffers
from heteroskedasticity problems. Consequently, to address this issue, it was deemed necessary
to re-estimate the model in logarithmic form.
Table 5 Logarithmic regression results
lgGDP
Coef.
St.Err.
t-
value
p-
value
[95%
Conf
Interval]
Sig
lgnumber
1.116
.046
24.38 0
1.026
1.207
***
Constant
7.734
.778
9.94
0
6.19
9.278
***
Mean dependent var
26.653
SD dependent var
1.520
R-squared
0.858
Number of obs
100
F-test
594.211
Prob > F
0.000
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 06,2025
Journal:
https://www.academicpublishers.org/journals/index.php/ijai
page 2205
Akaike crit. (AIC)
175.022
Bayesian crit. (BIC)
180.233
*** p<.01, ** p<.05, * p<.1
After transforming the variables and re-running the regression using the log-linear
specification, the Breusch-Pagan tests were conducted again to verify the presence of
heteroskedasticity. The results of these tests are reported in Table 6
.
Table 6. Brusch-Peagan heteroscedasticity test
Breusch-Pagan / Cook-Weisberg test for heteroskedasticity
Ho: Constant variance
Variables: fitted values of lgGDP
chi2(1)
= 3.77
Prob > chi2 = 0.0522
According to the results of the Bruce-Pagan test, it can be seen that the
heteroscedasticity problem in the model has been eliminated.
In the following table, we conduct the Bruce-Godfrey test to check for the
autocorrelation problem in the model.
Table 7. Brusch-Godfrey autocorrelation test
Number of gaps in sample:
6
Breusch-Godfrey LM test
for autocorrelation
chi2
df
Prob>Chi2
38.048
1
0.000
H0: no serial correlation
As can be seen from the results, our model has an autocorrelation problem. Since p<0.05,
it is concluded that the model has an autocorrelation problem. To overcome this problem, we
conduct the Niu-West autocorrelation-resistant standard errors test.
Table 8. Standard errors stable to Niue-West autocorrelation
GDP
Coef.
St.Err.
t-
value
p-
value
[95%
Conf
Interval]
Sig
number
14
412.266
16
25.826
8
.86
0
11
185.868
17
638.665
*
**
Consta
nt
1.
638e+11
4.
858e+10
3
.37
.
001
6.
739e+10
2.
602e+11
*
**
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 06,2025
Journal:
https://www.academicpublishers.org/journals/index.php/ijai
page 2206
Mean
dependent var
763977
662686.483
SD dependent
var
665217
206952.391
Number of obs
100
F-test
78.581
*** p<.01, ** p<.05, * p<.1
The results show that the autocorrelation problem in the model has been eliminated.
Conclusion
According to data from the World Travel & Tourism Council, in 2018, international
tourism contributed approximately USD 8.8 trillion to the global economy, accounting for
10.4% of total economic activity. Furthermore, the tourism sector was responsible for around
319 million jobs worldwide (Kun.uz, 2019) [3].
This study investigates the impact of the tourism industry on Gross Domestic Product
(GDP) in Turkey, Greece, Italy, Spain, Egypt, and Cyprus over the period from 2003 to 2022.
To analyze these relationships, a regression analysis was performed using the Ordinary Least
Squares (OLS) method in STATA-14. The results indicate that all regressors are statistically
significant, demonstrating that the development of the tourism sector positively contributes to
GDP growth.
Tourism plays a critical role in the economies of many countries and is essential for
national economic success. It increases revenue generation, creates numerous employment
opportunities, and fosters cultural exchange between foreign visitors and local citizens. In
addition to tourism expenditures, businesses and individuals often reinvest the income earned
from tourism-related activities, thereby significantly contributing to the development of local
economies.
References:
1. Wooldridge, J. M. (2020). Introductory Econometrics: A Modern Approach (7th ed.).
Cengage Learning.
2. Gujarati, D. N., & Porter, D. C. (2009). Basic Econometrics (5th ed.). McGraw-Hill
Education.
3. Cameron, A. C., & Trivedi, P. K. (2005). Microeconometrics: Methods and Applications.
Cambridge University Press.
4. STATA Corp. (2023). STATA Graphics Reference Manual. College Station, TX: StataCorp
LLC.
5. Chatfield, C. (2003). The Analysis of Time Series: An Introduction. CRC Press.
6. Greene, W. H. (2018). Econometric Analysis (8th ed.). Pearson Education.
7. Breusch, T. S., & Pagan, A. R. (1979). A simple test for heteroscedasticity and random
coefficient variation. Econometrica, 47(5), 1287–1294.
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 06,2025
Journal:
https://www.academicpublishers.org/journals/index.php/ijai
page 2207
8. World Travel & Tourism Council (WTTC). (2019). Economic Impact Report 2019. London:
WTTC.
9.
UNWTO. (2023). UNWTO Tourism Highlights. United Nations World Tourism
Organization.
10. Balaguer, J., & Cantavella-Jordá, M. (2002). Tourism as a long-run economic growth
factor: The Spanish case. Applied Economics, 34(7), 877–884.
11. Brida, J.G., Cortes-Jimenez, I., & Pulina, M. (2016). Has the tourism-led growth
hypothesis been validated? A literature review. Current Issues in Tourism, 19(5), 394–430.
12. Gunduz, L., & Hatemi-J, A. (2005). Is the tourism-led growth hypothesis valid for Turkey?
Applied Economics Letters, 12(8), 499–504.
13. Zorin, I. V., & Kvartalnov, V. A. (2004). Economics of Tourism. Moscow: Physical Culture
and Sports Publishing House. p. 35.
14. Bystrov, S. A., & Vorontsova, M. G. (2008). Tourism: Macroeconomics and
Microeconomics. Moscow–Saint Petersburg: Gerda Publishing House. p. 74.
15. Kun.uz. (2020). [Online] Available at: https://m.kun.uz/news/2020/05/24 [Accessed date,
e.g., 30 June 2025].
16. UNWTO. (2020). World Tourism Barometer, Vol. 18, Issue 1, January 2020.
17. Ibadullayev, N. E. (2010). Opportunities to Increase the Efficiency of Using Tourism
Resources (The Case of Samarkand Region). Ph.D. dissertation in Economics. Samarkand.
