Page 550
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
Macroeconomic Determinants of International Tourism Arrivals in
the Philippines
Joshua B. Magbago*, Melesa C. Cabusas
*
, & Adriane John P. Luncido*
Graduating Students, Bachelor of Arts in Economics, College of Arts and Sciences, Bukidnon State
University
DOI: https://doi.org/10.51583/IJLTEMAS.2026.150300045
Received: 11 March 2026; Accepted: 16 March 2026; Published: 09 April 2026
ABSTRACT
Tourism is a major driver of the Philippine economy, contributing significantly to employment, income
generation, and regional development. Despite its importance, empirical modeling of international tourism
demand in the Philippines remains limited. This study addresses this gap by estimating a tourism demand
function using two econometric approaches—Ordinary Least Squares (OLS) and Seemingly Unrelated
Regression (SUR)—to examine the macroeconomic determinants of international tourist arrivals. The analysis
focuses on the top five source markets—South Korea, the United States, Japan, China, and Australia—using
annual time-series data from 2007 to 2023. Data were sourced from the Department of Tourism, World Bank,
Philippine Statistics Authority, EM-DAT, and the Bangko Sentral ng Pilipinas. Descriptive results reveal a
steady increase in tourist arrivals from 2007 to 2019, followed by a sharp decline during the COVID-19
pandemic (2020–2021), and a gradual recovery beginning in 2022 with the easing of global travel restrictions.
Econometric findings indicate that real per capita income and exchange rates positively and statistically
significantly affect tourism demand, while relative prices and crime rates negatively affect arrivals. Notably, the
frequency of natural disasters exhibits a moderate positive association with tourism demand, possibly reflecting
post-disaster reconstruction efforts and intensified tourism promotion. Comparative results show that the SUR
model yields more efficient and consistent estimates than OLS, due to the presence of contemporaneous
correlation across country-specific equations. Overall, the findings underscore the interdependence of tourism
demand across major source markets and provide policy-relevant insights for enhancing the competitiveness and
resilience of the Philippine tourism sector.
Keywords: Contemporaneous correlation, demand, generalized least squares, seemingly unrelated regression
(SUR), tourism arrivals
INTRODUCTION
Tourism plays a pivotal role in driving economic growth by generating employment, creating income, and
supporting infrastructure development (Premovic & Arsic, 2020). In 2023, the global tourism industry supported
approximately 295 million jobs and contributed 9.1 percent to global GDP, equivalent to USD 9.5 trillion.
Despite persistent global economic uncertainties, the sector has demonstrated strong resilience (Theobald, 2012).
Given its significant socioeconomic contributions, sustained investments and sound policy interventions aimed
at enhancing visitor experience are essential for achieving inclusive and sustainable development. Although
international tourist arrivals have gradually recovered in recent years, the Philippines continues to lag behind
most ASEAN economies. In 2023, Thailand recorded over 28 million tourist arrivals, Vietnam received 12.06
million, while the Philippines registered only 5.45 million—an improvement from 2022 but still comparatively
low (The National Thailand, 2024).
_____________
1
Graduating Students, Bachelor of Arts in Economics, College of Arts and Sciences, Bukidnon State University, Malaybalay City, Bukidnon, Philippines
1
Faculty, Economics Department, College of Arts and Sciences, Bukidnon State University, Malaybalay City, Bukidnon, Philippines
Page 551
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
This gap underscores the need to examine the macroeconomic drivers of tourism performance through rigorous
econometric modeling to inform strategies that enhance recovery and competitiveness.
Existing empirical studies highlight the importance of macroeconomic and structural factors in shaping tourism
demand. For instance, Jeon (2014) found that income, market size, and direct flight availability positively
influence tourist arrivals, whereas geographical distance exerts a negative effect. Similarly, Manuela and de Vera
(2015) reported that aviation safety downgrades significantly reduce tourist inflows, while An (2022)
demonstrated that tourism demand determinants vary across country clusters. In addition, Camillo et al.
emphasized the combined influence of macroeconomic and environmental factors in explaining international
tourism flows.
The Philippines’ top five source markets are South Korea, the United States, Japan, China, and Australia
(Philippine News Agency, 2024). In 2022, the country recorded 2,010,522 international visitor arrivals, with
South Korea accounting for the largest share (27.19% or 546,726 arrivals), followed by the United States
(15.71%), China (6.49%), Japan (6.13%), and Australia (4.38%). These markets play a crucial role in shaping
aggregate tourism demand and therefore warrant focused empirical investigation.
Despite the growing body of literature, several gaps remain. First, existing studies largely emphasize cross-
country or regional analyses, with limited country-specific modeling for the Philippines. Second, much of the
literature relies on pre-pandemic data, failing to capture structural changes in travel behavior, income
responsiveness, and market dynamics following COVID-19. Third, prior research often examines individual
determinants in isolation, with limited integration of key macroeconomic variables—such as exchange rates,
inflation, and income—in a unified empirical framework. Addressing these gaps is essential for understanding
tourism demand in a post-pandemic, developing economy context.
This study contributes to the literature by providing a country-specific, post-pandemic analysis of the
macroeconomic determinants of international tourist arrivals in the Philippines. Using robust econometric
techniques, the study identifies key drivers of tourism demand across major source markets. The findings offer
evidence-based insights to support policy formulation aimed at strengthening tourism recovery, enhancing
resilience, and improving the country’s competitiveness within the ASEAN region.
Objectives of the Study
The primary objective of this study is to estimate the demand function for tourism arrivals in the Philippines.
Specifically, this research attempts to:
1) analyze the trends in international tourist arrivals from the Philippines’ top source markets from 2007
to 2023;
2) estimate the effects of key macroeconomic determinants on international tourist arrivals; and,
3) compare the performance and efficiency of Ordinary Least Squares (OLS) and Seemingly Unrelated
Regression (SUR) models in estimating tourism demand across source countries.
METHODOLOGY
Research Design
This study utilizes a descriptive–correlational research design to examine the relationships between
macroeconomic factors and international tourism arrivals in the Philippines. The descriptive component provides
an analysis of trends and patterns in tourist arrivals and selected macroeconomic indicators over the study period.
In contrast, the correlational component employs econometric modeling to estimate the magnitude and direction
of relationships between tourism demand and key explanatory variables, including real per capita income,
exchange rates, relative prices of goods and services, infrastructure expenditures, natural disasters, and crime
rates.
Page 552
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
This research design enables a systematic assessment of how macroeconomic conditions influence tourism
demand without manipulating the underlying variables, thereby preserving the observational nature of time-
series data (Creswell & Creswell, 2017). It is particularly appropriate for identifying statistically significant
associations and informing predictive insights in the context of tourism demand analysis.
Research Locale
The Philippines was chosen as the study location because it is a rich and diverse tourism destination. The country
is one of the best tourist destinations in ASEAN, comprising 16 regions, 82 provinces, and 7,641 islands
(Philippine Statistics Authority, 2023), which annually receive millions of visitors. The Philippines is known for
its cultural heritage, clean beaches, diving spots, and colorful festivals, yet it has achieved international
recognition. It received eight World Travel Awards in 2024, which underscores its growing role in the global
tourism market (GMA News, 2024).
Figure 1
The map of the Philippine islands
Note: Adapted from "Philippine Map High Resolution," by Emerson Carrie, 2020. Retrieved from Emerson
Carrie's Blog. Copyright 2020 by the author.
Data Analysis
Descriptive Statistics
This research has utilized secondary statistics from 2007 to 2023, from the Philippine Statistics Authority, EM-
DAT, Philippine Crime Rates Statistics, World Bank, Bangko Sentral ng Pilipinas, Department of Budget and
Management, and Department of Tourism to acquire the inbound tourists’ arrivals from South Korea, Australia,
China, the United States, and Japan.
Statistical Methods
To achieve the third objective of the study, multiple linear regression (MLR) and seemingly unrelated regression
(SUR) were used. These methods allow a comprehensive assessment of how independent variables, such as real
per capita income, the relative price of goods and services, exchange rates, infrastructure expenditures, natural
Page 553
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
disasters, and crime rates, affect tourism flows from South Korea, Australia, Singapore, the United States, and
Japan.
The Variables of the Study
Dependent Variable
International Tourist Arrivals from Top Five Source Countries refer to the annual number of inbound visitors to
the Philippines originating from its major tourism markets—South Korea, the United States, Japan, China, and
Australia. This variable serves as a proxy for tourism demand and reflects the overall attractiveness of the
Philippines as a destination. It is influenced by both domestic and international economic conditions and non-
economic factors, and is widely used as a key indicator of tourism inflows.
Independent Variables
Real Per Capita Income of the Origin Country
To obtain real GDP, nominal GDP is corrected for inflation by dividing it by the CPI. Real GDP is then divided
by the population to obtain real income per capita. An increase in real income typically implies greater
international mobility of residents; hence, this variable plays a significant role in the study of tourism flows to
the Philippines (Ongan et al., 2017). The formula to calculate the Real GDP per Capita of the origin country is:
Where: Y
it
= real income per capita of country i at time t
GDP
it
= Gross Domestic Product of the origin country at time t
CPI
it
= Consumer Price Index of country i at time t
Total Population
it
= Population of the country i at time t
Related Price
The related price (P) is determined by the ratio of the Philippine Consumer Price Index (CPI) to the CPI of the
origin country. This measure reflects the relative cost of goods and services between the two countries. A higher
P value indicates that the cost of living in the Philippines is higher than in the country of origin, which may
discourage inbound tourism due to higher travel and living costs. Conversely, a lower P value suggests that the
Philippines offers a more affordable destination for tourists, making it more attractive for international travellers
(Seetanah et al., 2015). The formula for the Related Price (P) between the Philippines and the origin country is:
Where: P
it
= relative price faced by tourists from the country at time
CPI
pht
= Consumer Price Index of the Philippines at time t
CPI
it
= Consumer Price Index of the origin country at time
The relative cost of tourism between a destination and a tourist-origin country is a key determinant of
international tourism demand. In the tourism demand literature, price competitiveness is frequently proxied by
relative consumer price levels because direct data on tourism-specific price indices are rarely available.
Consumer price indices (CPIs) are widely used as proxies for the general cost of goods and services encountered
by tourists in both destination and origin countries (Martin & Witt, 1987).
Page 554
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
While some studies incorporate exchange rate adjustments into price measures, this study models the exchange
rate as a separate explanatory variable to avoid potential multicollinearity and to isolate the distinct effects of
price levels and currency movements on tourism demand (Song & Witt, 2012).
Exchange Rate
The exchange rate is defined as the annual average bilateral nominal exchange rate, expressed as Philippine
pesos per unit of the origin country’s currency. It captures the effect of currency movements on tourists’
purchasing power. A depreciation of the Philippine peso (i.e., higher exchange rate) makes the destination
relatively cheaper for foreign visitors and is therefore expected to increase tourism demand.
Infrastructure Expenditures
Infrastructure expenditures refer to government spending on the development and maintenance of physical
infrastructure that supports tourism activity, including transportation networks (e.g., roads, airports, seaports),
utilities, and public services. This variable captures the extent to which a destination enhances accessibility,
connectivity, and service quality for visitors. Improved infrastructure reduces travel costs, increases
convenience, and enhances the overall tourist experience, thereby stimulating tourism demand.
Frequency of Natural Disasters
Frequency of natural disasters is defined as the annual number of recorded disaster events—such as typhoons,
floods, earthquakes, and other extreme natural hazards—affecting the destination country. This variable reflects
environmental risk and potential disruptions to tourism activity. A higher frequency of disasters can damage
infrastructure, disrupt transportation, and reduce destination attractiveness, thereby discouraging international
tourist arrivals.
Crime Rate
Crime rate refers to the number of reported criminal incidents within the destination country, typically measured
per 100,000 population per year. It serves as a proxy for the safety and security conditions tourists face. Higher
crime rates negatively influence destination image and perceived risk, which can deter both first-time and repeat
visitors, ultimately reducing tourism demand.
Econometric Models
The OLS method was used to estimate the model's parameters. This method yields the smallest sum of squared
vertical displacements between the observed values in the dataset and the regression model's estimates
(minimizing the sum of squared deviations from the estimated regression line). Assumptions, for the parameter
estimates to be BLUE (Best Linear Unbiased Estimator), are as follows:
a.)
= 0
This implies that the mean of the error terms is zero.
b.)
This is the property of homoscedasticity, meaning that the errors have a common variance.
c.)
where i ≠ j.
This is the property of no autocorrelation, meaning that no two error terms are serially correlated.
In this study, one of the key assumptions of the Ordinary Least Squares (OLS) method, specifically the
Page 555
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
assumption of no autocorrelation, was violated, justifying the use of alternative estimation techniques.
Having defined the variables, the full specification of the demand function of tourism in the Philippines is
presented in the Linear form as:
(1)
Where:
is the number of tourist arrivals from country i (i.e. South Korea, USA, Japan, China, and Australia) at
year t.
is the aggregate value of tourist arrivals in the Philippines when all other explanatory variables are equal
to zero. In many cases, β
o
has no clear economic interpretation, but it is always included in the model
because it helps in the overall estimation and prediction.
are the coefficients of all the explanatory variables for country i, which are parameters of how much
the change in value of the dependent variable when the value of an explanatory variable changes by its
unit, holding other variables constant.
represent the actual values of the five explanatory variables from country i at time t.
random error term of associated with the tourist arrivals from country i which are assumed to be normally
distributed, with a mean of zero and constant variance.
For this study, i represents the country where:
i = 1 for South Korea i = 4 for China
i = 2 for Japan i = 5 for Australia
i = 3 for the United States of America
Seemingly Unrelated Regression (SUR) Estimation
The seemingly unrelated regression (SUR) model (Zellner, 1962) comprises a set of equations in which the error
terms are correlated across equations. It is used when there are common factors that simultaneously influence
all equations, leading to cross-correlations in the errors. To enhance efficiency and accuracy in the study, the
SUR is applied to estimate tourism demand in South Korea, the USA, Japan, and Australia, which are the top
five source countries. The Breusch-Pagan test is used to test for contemporaneous correlation; a high test statistic
justifies the use of SUR rather than independent OLS models. The subscripts are used to represent variables and
countries (J, C, U, A, K) in the coefficients (). It is assumed that cross elasticities are zero. For the matrix
equations (1) to (5), they can be written compactly as follows:
(2)
Where
are (Tx1) vectors containing T observations on the logarithms of several tourist arrivals
Page 556
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
from South Korea, Japan, the USA, Singapore, and Australia, respectively. For this study, T=16, the total number
of observations for each country.
If we consider X for Japan, we have:
(Description for
Follow a similar description accordingly. The remaining matrices
and
can
be expanded as
The system of equations in (5) could further be written compactly as:
where:
The disturbance term is given by:
Hence, we can represent the four-equation “super model” in the environment of a single-equation linear model.
For SUR estimation, the disturbance variance is time-invariant within each equation or country, but each
equation may have a different variance (Judge et al., 1982).
However, the main characteristic of SUR is that disturbances across equations corresponding to the same period
are positively correlated. This is what we refer to as “contemporaneous correlation”. Wherever the disturbances
come in different periods, whether in the same or different equations, they are uncorrelated. Otherwise, we can
say that autocorrelation does not exist.
Testing for Contemporaneous Correlation
If contemporaneous correlation is absent, OLS for each equation, applied separately, is efficient in fullness, and
using the seemingly unrelated regression estimator is unnecessary. Thus, one can test whether the
contemporaneous covariances are zero. In this study, the null and alternative hypotheses for the
contemporaneous correlation test are:
Page 557
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
At least one covariance is non-zero.
Rejecting the null hypothesis implies there is a contemporaneous correlation and that the preference for SUR
over OLS is justified.
Treatment of Infrastructure Expenditure Variable
Infrastructure expenditure was initially included in the model as a potential determinant of tourism performance,
given its theoretical importance in supporting productive capacity, improving connectivity, and facilitating
economic growth. The variable was constructed using annual government infrastructure expenditure data for the
study period (17 observations). However, during the estimation of the Seemingly Unrelated Regression (SUR)
system, the infrastructure variable produced a coefficient estimate that was effectively zero across all equations.
This result showed that the variable did not have a measurable short-term effect on tourism arrivals within the
study period. In addition, the relatively small sample size and the possibility that infrastructure investment
influenced economic activity with longer time lags may have limited the model's ability to capture its short-run
effects.
Consequently, the infrastructure expenditure variable was omitted from the final SUR estimation to improve
model stability. The omission was not interpreted as evidence that infrastructure investment had no economic
impact; rather, it indicated that its effects were indirectly captured through other included variables or
materialized over longer horizons than those examined in the specification. This treatment ensured greater
transparency in the model specification and enhanced the robustness of the estimated results.
RESULTS AND DISCUSSION
Tourism Arrivals in the Philippines from Top Visiting Countries, 2007 – 2023
Tourism plays a significant role in the Philippines’ economic development, with international tourist arrivals
serving as a key indicator of the country’s performance and competitiveness in the global tourism market. Figure
2 presents the trends in tourist arrivals from the top five source markets—South Korea, the United States, Japan,
China, and Australia—over the period 2007 to 2019.
Figure 2 Trends of Tourist Arrivals in the Philippines, 2007-2023
Note: Tourist arrival trends in the Philippines from 2007 to 2023 are based on data adapted from the Department
of Tourism (DOT).
0
0.5
1
1.5
2
2.5
2007 2009 2011 2013 2015 2017 2019 2021 2023
Tourist Arrivals in Million
Year
South Korea United States of America Japan China Australia
Page 558
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
The data reveal a sustained upward trend in arrivals across all five countries prior to the COVID-19 pandemic,
indicating a period of expanding international demand for Philippine tourism. South Korea consistently remained
the largest source market, with arrivals reaching approximately 1.98 million in 2019. The United States ranked
second, exhibiting steady growth from 578,983 arrivals in 2007 to 1.62 million in 2018. China demonstrated
the fastest growth among the top markets, with arrivals peaking at 1.74 million in 2017, underscoring the growing
importance of emerging Asian markets. Japan followed with 682,788 arrivals in 2019, reflecting stable but
moderate growth over time. Meanwhile, Australia recorded the fewest arrivals among the five countries, though
it maintained a consistent upward trajectory—from 112,466 arrivals in 2007 to 286,170 in 2019. These trends
underscore the Philippine tourism sector's growing reliance on a concentrated set of major source markets,
particularly in Asia, prior to the pandemic. This pattern highlights both opportunities for market expansion and
potential vulnerabilities associated with market concentration.
Descriptive Statistics of the Variables
Tourist arrivals to the Philippines are primarily driven by South Korea, with an average of 914,000 visitors,
followed by the United States (750,000), China (446,990), Japan (388,820), and Australia, which records the
lowest average arrivals.
Table 1
Descriptive Statistics of the Variables
Variable
Average
Minimum
Maximum
Arrivals
South Korea
914520.00
6456.0
1607800.00
USA
758470.00
39326.00
1989300.00
Japan
388820.00
15024.0
682790.00
China
446990.00
9674.0
1743300.00
Australia
181950
2181
286170
Real Per Capita Income
South Korea
296034.09
278856.42
312316.17
USA
517.76
471.43
569.51
Japan
40575.40
38032.71
41753.75
China
481.19
324.94
638.50
Australia
42809.81
14469.27
107346.81
Related Price
South Korea
279852.35
172900
363130
USA
244.67
185.36
279.85
Japan
25356.82
5138
39004
China
1626.78
1378.3
1966.8
Australia
307.64
221.86
418.62
Page 559
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
Exchange Rate
South Korea
0.04
0.04
0.05
USA
47.76
42.22
55.63
Japan
0.54
0.38
1.95
China
7.17
6.05
8.09
Australia
38.47
34.14
44.72
Infrastructure Expenditure
305711049235
2314154000
893121040000
Natural Disaster
10.41
4
16
Crime Rates
8.15
4.32
10.84
Note: The Author’s calculation is based on the collected data.
This pattern indicates that South Korea and the United States serve as stable, high-volume source markets, while
China and Japan contribute moderately, and Australia remains a relatively underexploited market segment. In
terms of economic capacity, real per capita income is highest in South Korea (296,034.09), followed by Australia
(42,809.81) and Japan (40,575.40), while comparatively lower values are observed for China (481.19) and the
United States (517.76). These differences support the theoretical expectation that higher income levels enhance
the capacity for international travel (Adeleye et al., 2022). Relative price indicators suggest that the Philippines
is relatively more expensive for tourists from South Korea (279,852.35), Japan (25,356.82), and China
(1,626.78), while it remains more price-competitive for visitors from Australia (307.64) and the United States
(244.67). This finding is consistent with the notion that price competitiveness plays a critical role in destination
choice (Radukic et al., 2023). Exchange rate averages further highlight differences in purchasing power, with
higher values observed for the United States (47.76) and Australia (38.47), compared to China (7.17), Japan
(0.54), and South Korea (0.04). These variations suggest differing cost advantages for tourists across origin
countries, reinforcing the importance of currency movements in shaping tourism demand (Suh & Kim, 2018).
On the supply side, infrastructure expenditure averages 305.71 billion, reflecting sustained government
investment in facilities and services that support tourism development (Catudan, 2016). Meanwhile, the average
frequency of natural disasters is 10.41, indicating moderate exposure to environmental risks that may affect
tourism stability (Etulle-Tapanan, 2015). The average crime rate of 8.15 suggests moderate safety concerns,
which remain an important consideration in shaping tourists’ destination preferences.
OLS and Seemingly Unrelated Regression Results and Interpretation
A series of diagnostic tests was conducted to assess the validity of the Ordinary Least Squares (OLS)
assumptions. Given the time-series nature of the data, serial correlation is a primary concern. The Durbin–
Watson test results indicate positive autocorrelation across all five country-specific models, as reported in Table
2. This suggests a violation of the OLS assumption of no serial correlation, leading to the rejection of the null
hypothesis of independent error terms. In contrast, the results of the Breusch–Pagan, White’s, and Park tests
consistently show no evidence of heteroskedasticity across all models, indicating that the homoskedasticity
assumption is satisfied.
Table 2 Diagnostic Test Results for OLS Assumptions Across Country Models
Country
Autocorrelation (Durbin-
Watson)
Heteroskedasticity Tests
Positive
Negative
Breusch-Pagan-style test
White’s test
Park Test
South Korea
0.03
ns
0.96*
0.18*
0.18*
0.19*
Page 560
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
USA
0.00
ns
0.99*
0.10*
0.09*
0.14*
Japan
0.01
ns
0.99*
0.38*
0.34*
0.51*
China
0.00
ns
0.98*
0.10*
0.13*
0.19*
Australia
0.00
ns
0.99*
0.76*
0.60*
0.95*
* Satisfied
ns – not satisfied (Violated the Assumptions)
Given the presence of autocorrelation in the country-specific models, a test for contemporaneous correlation was
conducted. The Breusch–Pagan test results indicate statistically significant cross-equation error correlation (p-
value < 0.05), as reported in Table 3. This finding confirms that the error terms across the five country models
are interdependent, thereby justifying the use of Seemingly Unrelated Regression (SUR).
Table 3 Diagnostic Test Results for Seemingly Unrelated Regression (SUR)
Test Type
Test Statistic
Degrees of Freedom (df)
p-value
Breusch-Pagan LM Test for Diagonal
Covariance Matrix
101.94
10
0.0005
Table 4 presents the estimation results of Ordinary Least Squares (OLS) and Seemingly Unrelated Regression
(SUR) models examining the relationship between macroeconomic indicators and international tourist arrivals
to the Philippines from its primary source markets—South Korea, the United States, Japan, China, and Australia.
The dependent variable is tourist arrivals, while the explanatory variables include real per capita income, relative
price, exchange rate, infrastructure expenditure, frequency of natural disasters, and crime rate. OLS estimation
is limited in this context as it assumes independence across equations and fails to account for correlated
disturbances among countries. However, diagnostic results suggest the presence of both serial correlation and
contemporaneous correlation, likely arising from shared global shocks and common economic conditions
affecting multiple source markets simultaneously (Wooldridge, 2016; Gujarati, 2009). To address these
limitations, the study employs Seemingly Unrelated Regression (SUR), which explicitly models cross-equation
error correlations. Following Zellner's (1962) framework, SUR yields more efficient and consistent parameter
estimates than OLS, thereby enabling a more reliable assessment of the macroeconomic determinnts of tourism
demand across countries.
Table 4 OLS and SUR Estimates for Tourism Demand in the Philippines
Variable
Coefficient Estimates by Country
Ordinary Least Squares
Seemingly Unrelated Regression
South Korea
USA
Japan
China
Australia
South Korea
USA
Japan
Chin
a
Australia
INC
-16,152,000
ns
(19,980,000)
3,358,7
00
ns
(17,240
,000)
4,684,
600
ns
(4,239
,000)
58,759
ns
(114,9
00)
-11,301
ns
(107,300
)
12.85
ns
(14.71)
11,008
n
s
(8571)
107.5
4*
(36.3
1)
13,01
2*
(4,57
9)
-0.68
ns
(0.85)
Page 561
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
RP
3,166,900
ns
(3,370,000)
1,122,7
00
ns
(77070
00)
21,903
ns
(271,3
00)
122,84
0
ns
(128,0
00)
730,380
n
s
(752,600
)
6.17
ns
(4.33)
12,216*
(6,299)
19.77
**
(11.0
6)
-
3,504
.90**
(1,92
2)
1,854.5*
(630.1)
EX
5,551,200
ns
(6,440,000)
8,180,7
00
ns
(8,855,
000)
-
50,290
ns
(42,95
0)
-
871,36
0
ns
(492,0
00)
686,790
n
s
(1,098,00
0)
18,245,000
ns
(40,400,000)
81,841*
(35,800
)
438,8
60*
(186,
800)
-
777,2
80**
(454,
900)
8,678.4
ns
(7,802)
INF
-216.57
ns
(151,000)
176,450
ns
(102,00
0)
2337.7
ns
(44,46
0)
98,197
ns
(104,3
00)
8140.9
ns
(25,870)
0.00*
(0.00)
0.00*
(0.00)
0.00*
(0.00
)
0.00*
(0.00
)
0.00*
(0.00)
DIS
27,986
ns
(698,200)
107,430
ns
(379,90
0)
494,25
0
ns
(320,8
00)
231,53
0
ns
(338,1
00)
41,957
ns
(87,550)
33,428
ns
(36,910)
48,886*
*
(28,880
)
43,21
7*
(13,4
90)
47,25
6**
(25,3
10)
8,315.7
ns
(5,688)
CRI
-587,690
ns
(1,534,000)
-
4,052,9
00*
(1,257,
000)
-
570,99
0
ns
(483,8
00)
-
3,435,
900**
(792,8
00)
-
362,330
n
s
(231,000
)
77,927
ns
(68,430)
-
43,853
n
s
(59,500
)
-
2,933
.7
ns
(14,1
30)
-
67,11
4**
(38,2
00)
6,927.9
ns
(8,972)
_cons.
86,563,000
ns
(109,100,000)
-
21,233,
000
ns
(63,260
,000)
-
21,212
,000
ns
(19,83
0,000)
2,099,
400**
(104,3
00)
-
1,017,70
0
(2,213,00
0)
-5,980,300
ns
(4,836,000)
-
11,647,
000
**
(6,980,
000)
-
4,968
,800*
(1,53
5,000
)
5,845
,500
n
s
(3,69
4,000
)
-734560**
(443,000)
R
2
0.45
0.69
0.71
0.80
0.49
0.26
0.50
0.67
0.64
0.37
σ
2
418,160
323,690
124,19
0
267,38
0
73,799
372,820
316,010
100,3
80
270,8
40
62,971
System R
2
0.96
Page 562
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
Values in parentheses are the standard error of the coefficient
Note: INC is income INF is infrastructure expenditure
* Significant at 5%,
** Significant at 10%
RP is relative price
DIS is natural disaster of disaster
ns – not significant at the 5% and 10% level of significance
EX is the exchange rate
CRI is the crime rate
Interpretation of Estimates for Real Per Capita Income for the Philippines' Top Visiting Countries
Real per capita income captures the purchasing power of tourists in the origin country and is expected to
positively influence outbound travel. The SUR estimates indicate that income has a positive and statistically
significant effect on tourist arrivals from Japan (107.54) and China (13,012), while the coefficients for South
Korea (12.85), the United States (11,008), and Australia (-0.68) are statistically insignificant.
Under a linear specification, the coefficients imply that a one-unit increase in real per capita income is associated
with approximately 108 additional arrivals from Japan and 13,012 additional arrivals from China, holding other
factors constant. The absence of significance for South Korea and the United States suggests that tourism demand
from these mature markets may be less sensitive to income changes, possibly due to already high travel
propensity or saturation effects.
Consistent with tourism demand theory, these findings support the view that tourism is a normal good and, in
some cases, a luxury good, particularly in emerging markets (Lim et al., 2008). The findings support the view
that tourism demand responds to income growth, particularly in markets where outbound travel is still expanding.
Interpretation of Estimates for Relative Price for the Philippines' Top Visiting Countries
The relative price reflects the Philippines' cost competitiveness relative to the origin countries. The SUR results
show that the relative price is statistically significant for the United States (12,216), Japan (19.77), China (-
3,504.90), and Australia (1,854.5), but insignificant for South Korea.
The negative and significant coefficient for China is consistent with economic theory, indicating that an increase
in relative prices (i.e., the Philippines becoming more expensive) reduces tourist arrivals. Specifically, a one-
unit increase in relative price is associated with a decrease of approximately 3,505 Chinese tourists, ceteris
paribus.
In contrast, the positive coefficients for the United States, Japan, and Australia suggest that higher relative prices
are associated with increased arrivals. This counterintuitive result may reflect inelastic demand, strong
destination preferences, or the influence of omitted factors such as tourism promotion, diaspora ties, or travel
bundling (e.g., package tours), which may offset price effects.
These findings indicate that price sensitivity varies across markets, with China appearing more responsive to
cost conditions than other source countries.
Interpretation of Estimates for Infrastructure Expenditure for the Philippines' Top Visiting Countries
Infrastructure expenditure has a positive, statistically significant effect across all countries, though the magnitude
Page 563
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
is relatively small. In a linear framework, the coefficients indicate that increases in government spending on
infrastructure are associated with incremental increases in tourist arrivals.
This suggests that improvements in transportation, utilities, and tourism-related facilities enhance accessibility
and overall visitor experience, thereby supporting tourism demand. The consistency of significance across all
markets underscores the importance of sustained infrastructure investment in strengthening tourism
competitiveness.
Interpretation of Estimates for the Exchange Rate for the Philippines' Top Visiting Countries
The exchange rate influences tourism demand by affecting tourists’ purchasing power. The SUR estimates reveal
that exchange rate effects are positive and significant for the United States (81,841) and Japan (438,860). This
implies that a depreciation of the Philippine peso (i.e., an increase in the exchange rate) makes the destination
more affordable and is associated with higher tourist arrivals.
In contrast, the coefficient for China (-777,280) is negative and significant, indicating that peso depreciation is
associated with a decline in Chinese arrivals. This result suggests that exchange rate effects are market-specific
and may reflect differences in travel behavior, policy constraints, or substitution patterns across destinations.
The coefficients for South Korea and Australia are not statistically significant, suggesting that exchange rate
movements do not decisively influence tourism demand from these countries.
These findings are consistent with prior studies emphasizing that exchange rate effects are country-specific and
context-dependent (Yap, 2013).
Interpretation of Estimates for the Natural Disaster for the Philippines' Top Visiting Countries
Natural disasters—such as typhoons, earthquakes, and floods—are generally expected to reduce tourism demand
due to travel disruptions and infrastructure damage (Rosselló et al., 2020; Ritchie, 2004). However, the SUR
estimates indicate positive and statistically significant coefficients for the United States (48,886), Japan (43,217),
and China (47,256), while the effects for South Korea (33,428) and Australia (8,315.7) are statistically
insignificant.
In a linear specification, these estimates imply that an additional recorded disaster event within a year is
associated with higher arrivals from the United States, Japan, and China, holding other factors constant. This
counterintuitive result should not be interpreted as disasters promoting tourism. Rather, it likely reflects temporal
aggregation effects and demand resilience—that is, annual tourist arrivals may remain strong or recover within
the same period despite short-term disruptions.
Several mechanisms may explain this pattern. First, the use of annual data may mask short-lived declines
immediately after disaster events, as recovery and reconstruction activities, as well as intensified tourism
promotion, can offset initial losses within the year. Second, major source markets may exhibit low sensitivity to
short-term shocks, particularly when travel plans are pre-booked or when alternative destinations are limited.
Third, the Philippines’ tourism sector may demonstrate some structural resilience, allowing it to absorb and
recover from environmental shocks without substantial long-term declines in arrivals.
The statistically insignificant results for South Korea and Australia further suggest that natural disasters do not
exert a measurable effect on tourism demand from these markets. This may reflect differences in risk perception,
travel behavior, or information processing across nationalities (Reisinger & Mavondo, 2005).
The findings indicate that the impact of natural disasters on tourism demand is heterogeneous and context-
dependent, and that their adverse effects may be short-term and not fully captured in annual aggregate data.
Interpretation of Estimates for the Crime Rates for the Philippines' Top Visiting Countries
Crime rates are an important determinant of tourism demand, as they shape perceived risk and destination image,
Page 564
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
which in turn influence travel decisions. The SUR estimates indicate that crime has a negative, statistically
significant effect only in China (−67,114), while the coefficients for the United States (−43,853), Japan (2,933.7),
South Korea (77,927), and Australia (6,927.9) are statistically insignificant.
In a linear framework, this implies that an increase in crime is associated with a reduction in tourist arrivals from
China, holding other factors constant. In contrast, the absence of statistical significance for the other countries
suggests that crime does not exert a measurable effect on tourism demand from these markets during the study
period.
These findings are consistent with the tourism risk literature, which emphasizes that perceived risk, rather than
objective crime levels alone, drives tourist behavior (Reisinger & Mavondo, 2005; Karl, 2018). Empirical
evidence further shows that safety perception plays a critical role in shaping destination image and travel
intention, with higher perceived insecurity reducing tourism demand (Ding & Wu, 2022).
The cross-country differences observed in the results highlight heterogeneity in risk sensitivity across source
markets. Tourists from the United States, Japan, South Korea, and Australia may be less responsive to marginal
changes in crime due to risk-mitigating strategies, such as organized travel, pre-arranged itineraries, and reliance
on official travel advisories. Additionally, these tourists may already incorporate baseline safety conditions into
their destination choices, making short-term fluctuations in crime less influential.
In contrast, the significant negative response among Chinese tourists may reflect greater sensitivity to safety
concerns or stronger influence from external information channels, such as media coverage and travel networks.
From a theoretical perspective, this behavior is consistent with expected utility theory, wherein tourists evaluate
destinations by weighing expected benefits against perceived risks.
Test for Equality of Coefficients Per Country
To further validate the results, a Wald test was conducted to assess whether the estimated coefficients of each
explanatory variable are statistically equal across the five source-country equations within the Seemingly
Unrelated Regression (SUR) framework. Specifically, the test evaluates the null hypothesis that the coefficients
for a given variable (e.g., real per capita income, relative price, exchange rate) are identical across all countries—
South Korea, the United States, Japan, China, and Australia.
As reported in Table 5, the Wald test provides a formal basis for determining whether macroeconomic
determinants exert homogeneous or heterogeneous effects on international tourist arrivals. Rejection of the null
hypothesis indicates that the impact of a given variable differs significantly across source markets, thereby
supporting the presence of cross-country heterogeneity in tourism demand.
Table 5 Wald Tests for Equality of Coefficients Across SUR Equations
Test of Parameter Equality Across Equations
Wald Chi-Square Statistic
P-Value
INC South Korea=USA=China=Japan=Australia
19.438*
0.000
RP South Korea=USA=China=Japan=Australia
16.996*
0.001
EX South Korea=USA=China=Japan=Australia
12.419*
0.014
INF South Korea=USA=China=Japan=Australia
6.280
ns
0.179
DIS South Korea=USA=China=Japan=Australia
8.936**
0.063
CRI South Korea=USA=China=Japan=Australia
5.519
ns
0.238
* Significant at 5%, ** Significant at 10%
Page 565
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
ns – not significant at the 5% level of significance
The Wald tests for equality of coefficients across the SUR equations reveal significant cross-country differences
in how macroeconomic determinants influence international tourist arrivals to the Philippines. At the 5 percent
significance level, the null hypothesis of equal coefficients is rejected for real per capita income, relative price,
and exchange rate, indicating that these variables exert heterogeneous effects across the five source markets.
This implies that tourists from South Korea, the United States, China, Japan, and Australia respond differently
to changes in income, cost conditions, and currency movements. For instance, the results suggest that the
sensitivity of tourism demand to income and exchange rate fluctuations varies across countries, reflecting
differences in purchasing power, travel behavior, and market characteristics.
Similarly, the effects of infrastructure expenditure are not uniform across countries, indicating that
improvements in tourism-related facilities—such as airports, transportation systems, and accommodations—
benefit source markets to varying degrees. In contrast, natural disasters exhibit weaker cross-country differences,
with the null hypothesis rejected only at the 10 percent level, suggesting more limited heterogeneity in their
impact. Finally, the Wald test indicates that crime rates do not differ significantly across countries. This suggests
that the effect of crime on tourism demand is relatively uniform across source markets, although this should not
be interpreted as evidence that all tourists perceive the Philippines as uniformly safe. Rather, it indicates that
cross-country differences in responsiveness to crime are not statistically distinguishable within the model.
Testing for Contemporaneous Correlation
To test whether or not there is a reason to prefer SUR over OLS (i.e., whether contemporaneous correlation exists
or not), the study proceeded with the following hypothesis:
At least one covariance is non-zero.
The test statistic given as a Lagrange Multiplier statistic, suggested by Breusch and Pagan (1980), is written as:
or
Given the OLS residual covariance matrix, which gives the variances (diagonal) and covariances (off-
diagonal) to compute each residual correlation, written as:
Where under
Has an x
2
distribution with 3 degrees of freedom (In general, the degrees of freedom are
equal to N(N-1) where N is the number of equations.)
The test obtained
= 17(0.4242955 + 0.2839086 + 0.3732883 + 0.7371199 + 0.5197286 + 0.9101933 + 0.8433128 + 0.6143392
+ 0.5481716 + 0.7424745)
= 17 (5.99683)
= 101.94
With a critical value of x
2
(0.05,10)
≈ 18.31. Since 101.94 > 18.31, the null hypothesis of no contemporaneous
correlation is rejected. Rejecting the null hypothesis implies there is contemporaneous correlation; hence, SUR
is preferred over separate OLS.
The value of the Breusch-Pagan Lagrange Multiplier (LM) test is 101.94, which is considerably greater than the
Page 566
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
critical chi-square value of 18.31 under 10 degrees of freedom. The result of the Breusch Pagan test shows that
there is a high contemporaneous correlation between the five countries, and this indicates that tourism behaviour
is interdependent in the five countries. This implies that changes in the arrival of tourists in one market are likely
to be accompanied by changes in other markets. There are global shocks, such as economic downturns or
pandemics, that affect all countries simultaneously. But the closest connection is between South Korea and
China, which is reactive to changes in price and exchange rates. The moderate linkages among the U.S., Japan,
and Australia are also evident in perceptions of cost and safety. These tendencies align with the tourism
interdependence theory, which emphasizes the impact of the global environment on travel demand. Thus, the
SUR model should be used to analyze multiple countries, with the null hypothesis rejected.
LIMITATIONS OF THE STUDY
Despite its contributions, this study is subject to several limitations. First, the analysis relies on a relatively small
time-series dataset spanning 2007 to 2023. While this period captures important events, including the COVID-
19 pandemic, the limited number of observations may reduce the statistical power of the estimates and constrain
the robustness of inference.
Second, the model focuses primarily on macroeconomic determinants and does not account for other relevant
factors that may influence tourism demand, such as air connectivity, visa policies, tourism marketing
expenditure, and digital promotion. Excluding these variables may lead to omitted-variable bias, potentially
affecting the estimated relationships.
Third, the presence of major structural shocks—most notably the COVID-19 pandemic—may introduce
structural breaks in the time series that are not fully captured within the current modeling framework. As a result,
the estimated relationships may not fully reflect changes in tourism behavior during periods of disruption.
Fourth, although the Seemingly Unrelated Regression (SUR) approach accounts for contemporaneous cross-
country correlation, it does not explicitly model dynamic adjustments or long-run equilibrium relationships,
which may be important in tourism demand analysis.
Future research may address these limitations by incorporating a broader set of explanatory variables, utilizing
higher-frequency data (e.g., quarterly or monthly), and applying dynamic econometric techniques—such as
Autoregressive Distributed Lag (ARDL) or Vector Error Correction Models (VECM)—to better capture both
short-run dynamics and long-run relationships in tourism demand.
CONCLUSION
This study demonstrates that key macroeconomic variables play an important role in explaining international
tourist arrivals to the Philippines. The application of Seemingly Unrelated Regression (SUR) yields more
efficient and reliable estimates than Ordinary Least Squares (OLS), given the presence of contemporaneous
correlation across country-specific equations.
The empirical results reveal that real per capita income significantly influences tourist arrivals from Japan and
China, indicating that income-driven demand is more pronounced in these markets. The relative price is found
to be a significant determinant in Australia, Japan, and the United States, highlighting the role of cost
competitiveness in shaping tourism flows. Meanwhile, the exchange rate exerts a positive and significant effect
on arrivals from the United States and Japan, suggesting that currency depreciation enhances the Philippines’
attractiveness as a destination.
In addition, infrastructure development shows a generally positive association with tourism demand,
underscoring the importance of sustained investment in tourism-supporting facilities. The results for natural
disasters indicate that their impact is not uniformly negative; while disruptions may occur in the short run,
tourism demand appears resilient over time, reflecting recovery dynamics and adaptive capacity. Crime rate, on
the other hand, negatively affects tourism demand, though its impact varies across source markets.
Page 567
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
This study confirms that tourism demand in the Philippines is sensitive to macroeconomic conditions and that
these effects are heterogeneous across major source countries. The presence of contemporaneous correlation
further highlights the interconnected nature of tourism demand across markets, reinforcing the suitability of the
SUR framework for this analysis.
RECOMMENDATIONS
Based on the findings of this study, several policy and research recommendations are proposed to enhance the
performance and resilience of the Philippine tourism industry.
First, the Department of Tourism, in coordination with relevant government agencies and private sector
stakeholders, should strengthen the country’s price competitiveness. This may be achieved through the
development of affordable yet high-quality tourism packages, the promotion of bundled services, and the
provision of targeted discounts or incentives during periods of economic slowdown. Such measures can help
sustain demand, particularly in price-sensitive markets.
Second, the government should prioritize sustained investments in tourism-supporting infrastructure, including
modernizing airports, improving transport connectivity, and upgrading digital and public service facilities.
Enhancing infrastructure is essential for improving accessibility, service quality, and overall visitor experience,
thereby increasing the country’s attractiveness as a tourism destination.
Third, ensuring safety and security remains critical. Strengthening law enforcement, improving tourist protection
mechanisms, and implementing effective communication strategies that promote the Philippines as a safe
destination are essential in maintaining positive destination image and visitor confidence.
Fourth, policymakers should adopt a market-specific approach to tourism development, recognizing that source
countries respond differently to macroeconomic conditions such as income, prices, and exchange rates. Tailored
marketing strategies and targeted interventions can therefore be designed to better align with each market's
preferences and sensitivities.
Finally, future research should expand the scope of analysis by incorporating shock-sensitive and structural
variables, such as pandemics, geopolitical risks, flight connectivity, visa facilitation, tourism promotion, and
digital infrastructure. Researchers are also encouraged to adopt dynamic econometric approaches, including
Autoregressive Distributed Lag (ARDL) and Vector Error Correction Models (VECM), to capture both short-
run adjustments and long-run relationships in tourism demand. In addition, the use of primary data collection
methods—such as surveys, interviews, or digital tracking—can provide deeper insights into tourist behavior,
preferences, and spending patterns.
REFERENCES
1. An, F. (2022). Exploring influencing factors on international tourist arrivals in the Philippines using
clustering methods and negative binomial regression. Retrieved from
https://animorepository.dlsu.edu.ph/etdm_math/6
2. Becker, G. S. (1968). Crime and punishment: An economic approach. Journal of political
economy, 76(2), 169-217.Carrie, E. (2020). Philippine map high resolution.
https://i.pinimg.com/736x/9c/1b/f9/9c1bf95cf2029268bcdd14cd052a02f9.jpg
3. Creswell, J. W., & Creswell, J. D. (2017). Research Design: Qualitative, Quantitative, And Mixed
Methods Approaches. Sage Publications.
4. Philippine News Agency. (2024, April 25). PH logs over 2M int’l visitors – DOT.
https://www.pna.gov.ph/articles/1223391
5. Ding, G., & Wu, J. (2022). Influence of tourism safety perception on destination image: A case study of
Xinjiang, China. Sustainability, 14(3), 1663.
6. GMA News. (2024, September 4). The Philippines Wins Eight Titles At The World Travel Awards 2024.
GMA Network. https://www.gmanetwork.com/news/lifestyle/travel/919235/philippines-wins-8-titles-
at-world-travel-awards-2024/story/
Page 568
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue III, March 2026
7. Gujarati, D. N., & Porter, D. C. (2009). Basic econometrics. McGraw-hill.
8. Jeon, N. (2014). Determinants Of International Tourism Demand For The Philippines: An Augmented
Gravity Model Approach.
9. Judge, G. G. (1982). Introduction To The Theory And Practice Of Econometrics (Illustrated Ed.). Wiley.
10. Karl, M. (2018). Risk and uncertainty in travel decision-making: Tourist and destination
perspective. Journal of Travel Research, 57(1), 129-146.
11. Lim, C., Min, J. C., & McAleer, M. (2008). Modelling income effects on long and short haul international
travel from Japan. Tourism Management, 29(6), 1099-1109.
12. Manuela Jr, W. S., & de Vera, M. J. (2015). The impact of government failure on tourism in the
Philippines. Transport Policy, 43, 11–22.
13. Nguyen, L. P., & Nguyen, H. T. (2021). Factors impacting tourism demand: An analysis of 10 ASEAN
countries. The Journal of Asian Finance, Economics, and Business, 8(1), 385–393.
14. Ongan, S., Isik, C., & Özdemir, D. (2017). The effects of real exchange rates and income on international
tourism demand for the USA from some European Union countries. Economies,
https://doi.org/10.3390/economies5040051 5(4), 51.
15. Philippine Statistics Authority (PSA). (2023). Philippine Population And Administrative Divisions.
https://www.foi.gov.ph/requests/the-total-population-by-region-in-the-philippines-every-year-2010-
2023/
16. Premović, J., & Arsić, L. (2020). Socio-Economic Aspects of Tourism in the Modern Society. Зборник
радова Филозофског факултета у Приштини, 50(3), 125-154.
17. Reiner, V., Pathirana, N. L., Sun, Y. Y., Lenzen, M., & Malik, A. (2024). Wish you were here? The
economic impact of the tourism shutdown from australia’s 2019-20 ‘black summer’bushfires. Economics
of Disasters and Climate Change, 8(1), 107-127.
18. Reisinger, Y., & Mavondo, F. (2005). Travel anxiety and intentions to travel internationally: Implications
of travel risk perception. Journal of travel research, 43(3), 212-225.
19. Song, H., & Witt, S. F. (2012). Tourism Demand Modelling And Forecasting. Routledge.
20. Suh, H., & Kim, S. B. (2018). The macroeconomic determinants of international casino travel: Evidence
from South Korea’s top four inbound markets. Sustainability, 10(2), 554.
21. The Nation Thailand. (2024, January 27). Thailand tops ASEAN tourism market, with 28 million-plus
arrivals in 2023. The Nation Thailand.
https://www.nationthailand.com/thailand/tourism/40035063
22. Theobald, W. F. (Ed.). (2012). Global Tourism. Routledge.
23. Witt, S. F., & Martin, C. A. (1987). Tourism demand forecasting models: Choice of an appropriate
variable to represent tourists’ cost of living. Tourism Management, 8(3), 233-246.
24. Wooldridge, J. M. (2016). Introductory Econometrics A Modern Approach. South-Western cengage
learning.
25. Yap, G. (2013). The impacts of exchange rates on Australia's domestic and outbound travel
markets. Mathematics and computers in simulation, 93, 139-150.
26. Zellner, A. (1962). An efficient method of estimating seemingly unrelated regressions and tests for
aggregation bias. Journal of the American Statistical Association, 57(298), 348–368.
