ArticlePDF Available

Abstract

A thorough understanding of the fluctuations of tourist flows provides useful insights concerning the nature of tourist demand. This study aims to investigate the fluctuation patterns and dynamics of inbound tourist flows to China using a complex network approach. Several measures, such as the network topological parameters of degree and degree distribution, betweenness centrality, and shortest path length, are utilized to discover important fluctuation patterns and the transition distance. Based on the empirical results, six important fluctuation patterns of inbound tourist flows to China are recognized. These fluctuation patterns are important intermediaries in the process of transformation of the fluctuation patterns and can be viewed as a prelude to changes in the inbound tourist flows. The value of 3.38 found for the average transition distance suggests that the transformation occurs approximately every three to four quarters. These findings are useful for understanding the inherent laws and transformations governing fluctuations in tourist flows.
This article was downloaded by: [Stephen F Austin State University]
On: 05 August 2015, At: 04:23
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: 5
Howick Place, London, SW1P 1WG
Click for updates
Asia Pacific Journal of Tourism Research
Publication details, including instructions for authors and subscription
information:
http://www.tandfonline.com/loi/rapt20
Modeling the Fluctuation Patterns of Monthly
Inbound Tourist Flows to China: A Complex
Network Approach
Yongrui Guoa, Jie Zhanga, Yang Yangb & Honglei Zhanga
a Department of Land Resources and Tourism Sciences, Nanjing University,
Nanjing, Jiangsu, People's Republic of China
b School of Tourism and Hospitality Management, Temple University,
Philadelphia, PA, USA
Published online: 26 Aug 2014.
To cite this article: Yongrui Guo, Jie Zhang, Yang Yang & Honglei Zhang (2015) Modeling the Fluctuation
Patterns of Monthly Inbound Tourist Flows to China: A Complex Network Approach, Asia Pacific Journal of
Tourism Research, 20:8, 942-953, DOI: 10.1080/10941665.2014.948024
To link to this article: http://dx.doi.org/10.1080/10941665.2014.948024
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”)
contained in the publications on our platform. However, Taylor & Francis, our agents, and our
licensors make no representations or warranties whatsoever as to the accuracy, completeness, or
suitability for any purpose of the Content. Any opinions and views expressed in this publication
are the opinions and views of the authors, and are not the views of or endorsed by Taylor &
Francis. The accuracy of the Content should not be relied upon and should be independently
verified with primary sources of information. Taylor and Francis shall not be liable for any
losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities
whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial
or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or
distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use
can be found at http://www.tandfonline.com/page/terms-and-conditions
Modeling the Fluctuation Patterns of Monthly
Inbound Tourist Flows to China: A Complex
Network Approach
Yongrui Guo
1
, Jie Zhang
1
, Yang Yang
2
and Honglei Zhang
1
1
Department of Land Resources and Tourism Sciences, Nanjing University, Nanjing, Jiangsu,
People’s Republic of China
2
School of Tourism and Hospitality Management, Temple University, Philadelphia, PA, USA
A thorough understanding of the fluctuations of tourist flows provides useful insights con-
cerning the nature of tourist demand. This study aims to investigate the fluctuation pat-
terns and dynamics of inbound tourist flows to China using a complex network
approach. Several measures, such as the network topological parameters of degree and
degree distribution, betweenness centrality, and shortest path length, are utilized to dis-
cover important fluctuation patterns and the transition distance. Based on the empirical
results, six important fluctuation patterns of inbound tourist flows to China are recog-
nized. These fluctuation patterns are important intermediaries in the process of transform-
ation of the fluctuation patterns and can be viewed as a prelude to changes in the inbound
tourist flows. The value of 3.38 found for the average transition distance suggests that the
transformation occurs approximately every three to four quarters. These findings are
useful for understanding the inherent laws and transformations governing fluctuations
in tourist flows.
Key words: inbound tourist, tourist flows, fluctuation patterns, complex network, China
Introduction
Since China adopted the “open-door” policy
in 1978, tourism, especially inbound tourism,
has developed rapidly (Yang & Wong,
2013). In 1986, tourism, as an economic
industry, was incorporated for the first time
into the five-year state plan for the National
Economy and Social Development; in 1992,
tourism was included as one of the key indus-
tries in the tertiary sector; in 1998, tourism
was selected as a new growth pole of the
national economy; and in 2009, the State
Council of the People’s Republic of China
Asia Pacific Journal of Tourism Research, 2015
Vol. 20, No. 8, 942953, http://dx.doi.org/10.1080/10941665.2014.948024
Corresponding author. Email: zhanghonglei@nju.edu.cn
#2014 Asia Pacific Tourism Association
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
issued a directive to upgrade tourism to a stra-
tegic pillar industry in the national economy.
Inbound tourist flows to China increased from
27.46 million in 1990 to 132.40 million in
2012, representing an annual growth rate of
more than 7.4% (CNTA, 2013). Inbound
tourism, with its important role in securing
foreign exchange earnings, is of great impor-
tance to the economy of China (Yang &
Wong, 2013). Inbound tourism receipts
increased from US$2,217.58 million in 1990
to US$50,028.00 million in 2012 (CNTA,
2013). In view of the rapid increase in
inbound tourist flows and receipts over the
past few decades, a comprehensive examination
of fluctuations of inbound tourist flows is of
importance to both tourism business prac-
titioners and tourism policy-makers. In this
article, we introduce a complex network
approach to model the fluctuation patterns of
monthly inbound tourist flows to China.
Temporal fluctuations of tourist flows trig-
gered by seasonality and business cycles are
one of the most significant characteristics of
tourism (Assaf, Barros, & Gil-Alana, 2011;
Cuccia & Rizzo, 2011; Fourie & Santana-
Gallego, 2011;Nadal,Font,&Rossello,
2004; Song & Li, 2008;Vergori,2012). Over
the past three decades, many studies on
tourism demand analysis and forecasting have
contributed significantly to our understanding
of the temporal fluctuations of tourism
demand (Song, Li, Witt, & Athanasopoulos,
2011). However, although the importance of
temporal fluctuations of tourist flows has been
broadly recognized, it has also been acknowl-
edged that this phenomenon is not well under-
stood (De Cantis, Ferrante, & Vaccina, 2011;
Higham & Hinch, 2002;Jang,2004;Lim&
McAleer, 2001). Most existing studies of
tourism demand have involved a forecasting
perspective, and fewer studies have focused
on the fluctuation patterns of tourist flows
(Chan, Lim, & McAleer, 2005;Cho,2009).
An in-depth understanding of the fluctuations
of tourist flows is the basis for accurate predic-
tions of future trends in tourism demand. It is
important to scrutinize changes in both the
patterns and the amplitude of fluctuations (De
Cantis et al., 2011). The aim of this study is
to examine the fluctuation patterns, amplitude,
and dynamics of monthly inbound tourist flows
to China from January 1990 to December 2012
using a complex network approach.
The paper is organized as follows. After this
introductory section, we briefly outline the
recent literature relating to the study of
tourism demand. The subsequent section
describes our methodology and data source.
Using a complex network approach, we
analyze the fluctuation patterns, amplitude,
and dynamics of inbound tourist flows to
China. The results are presented in the next
section, and some concluding remarks are
offered in the final section.
Literature Review
Tourism demand forecasting is one of the most
important task for planning development and
operational management in the tourism indus-
try. Exploring the fluctuation patterns and
accurately forecasting the future tourist flows
are essential to determine successful invest-
ments for both the public and the private
sectors (Chang & Liao, 2010). The information
from these investigations and forecasts plays a
highly important role in formulating national
tourism development policy and strategic plan-
ning, optimizing allocation of tourism market
resources, and conducting decision-making for
tourism businesses (Tao & Ni, 2010).
Measuring and analyzing fluctuation is an
important aspect of the study of tourism
demand (Chu, Yeh, & Chang, 2014;Turner
Modeling the Fluctuations of Tourist Flows 943
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
&Witt,2001). Gil-Alana (2005) examined
monthly international tourist flows in the USA
by assuming seasonal univariate long-memory
processes and suggested that the total number
of arrivals implied long memory and mean
reverting behavior. Shareef and McAleer
(2005) analyzed the conditional volatility of
tourist flows to small island tourism economies
and found that the logarithm of monthly inter-
national tourist arrivals was stationary. Yan
and Wall (2003) identified the structure, charac-
teristics, and intensity of fluctuations in the
number of international visitors to China from
1980 to 1998 and showed that the overall
trend was strong annual growth, influenced by
a cyclical fluctuation. Several studies have
applied sophisticated methods for forecasting
tourism demand, such as the space– time
cluster approach (Gursoy, Parroco, & Scuderi,
2013), multivariate exponential smoothing
(Athanasopoulos & de Silva, 2012), evolution-
ary fuzzy systems (Hadavandi, Ghanbari, Sha-
hanaghi, & Abbasian-Naghneh, 2011), the
autoregressive integrated moving average
model (Coshall, 2006; Gustavsson & Nord-
strom, 2001), dynamic almost ideal demand
system approach (Kuo, Liu, & Chen, 2014),
the seasonal autoregressive integrated moving
average model (Goh & Law, 2002), the vector
autoregressive model (Song & Witt, 2006), the
autoregressive distributed lag model (Song,
Lin, Zhang, & Gao, 2010), the Lagrange multi-
plier unit root tests (Lean & Smyth, 2009), and
time-varying parameter error correction model
(Li, Wong, Song, & Witt, 2006). Several
studies have also compared the forecasting
accuracy of various models and approaches
(Cho, 2003; Kim & Moosa, 2001; Wong,
Song, Witt, & Wu, 2007). Two extensive
reviews in this area are available from Song
and Li (2008) and Li, Song, and Witt (2005).
Identification of the fluctuation character-
istics of time series data is of crucial impor-
tance in a wide variety of fields. Many
methods, such as the Lyapunov exponent,
the autoregressive conditional heteroscedasti-
city model, and the stochastic volatility
model, have been used to analyze these charac-
teristics. These traditional methods focus
primarily on the overall features of the time
series but cannot provide nuanced information
on the determinations of the system properties
(Yang, Pan, & Song, 2014;Zhang,Zhou,
Jiang, & Wang, 2010). The study of time
series data using a complex network approach
has attracted great interests among scholars.
Time series can be mapped as a complex
network using various methods, such as the
visibility graph algorithm (Lacasa, Luque, Bal-
lesteros, Luque, & Nuno, 2008) and the
coarse-graining process (Li & Wang, 2007).
Through the application of the complex
network approach, the temporal dynamics of
time series data is encoded into the topology
of the corresponding networks. According to
the statistical properties of the network, the
determinations of different fluctuation patterns
can be identified (Zhang et al., 2010). In finan-
cial time series, for example, Bonanno et al.
(2004)showed that a network can be obtained
by a correlation-based filtering procedure and
that meaningful economic information can be
extracted from noise-dressed correlation
matrices. Fluctuations or temporal imbalances
in tourist flows create an evolving complex
dynamic system. In nature and society, many
complex dynamic systems can be represented
as complex networks (Li & Wang, 2007).
Complex network approach offers a promising
new method to the analysis of tourist flow time
series data.
In this paper, we develop a weighted
network of monthly inbound tourist flows to
China from January 1990 to December
2012. The network can translate inbound
tourist flows to various characteristics in its
944 Yongrui Guo et al.
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
network topological structure. Every node
in the network corresponds to a distinct
fluctuation pattern and has a special role in
shaping the dynamics of inbound tourist
flows. We introduce several effective par-
ameters for detecting important topological
nodes of the network of inbound tourist
flows. From these nodes, we can obtain signifi-
cant fluctuation patterns of inbound tourist
flows to China. We focus on the fluctuations
and correlations of changes in inbound
tourist flows. The statistical properties of fluc-
tuations in inbound tourist flows are impor-
tant for understanding and modeling the
complex dynamics of inbound tourist flows.
Data Source and Methodology
Data Source
The monthly time series data for inbound
tourist flows to China used in the present
study were obtained from The yearbook of
China tourism statistics (19912013). The
series contains 276 data points from January
1990 to December 2012 (Figure 1). The term
“inbound tourists” to China refers to foreign
tourists and tourists from Hong Kong,
Macau, and Taiwan. The time series of
inbound tourist flows to China was rep-
resented as T(t)(t¼1, 2, 3, ...,N,N¼276).
Coarse-Graining Preprocess
The simplest possible method for transforming
a time series into a complex network represen-
tation is to coarse grain its range into a suitable
set of classes and to consider the transition
probabilities between these classes in terms
of a weighted network. The coarse-graining
process is an effective method for analyzing
the complexity of time series data. After a
time series interval has been divided into
homogeneous partitions, the interval can be
averaged into limited subintervals (Li &
Wang, 2007). By giving each subinterval a
symbol, the time series data can be represented
as a discrete symbolic sequence; studying the
time series is then equivalent to studying
the symbolic sequence. The coarse-graining
process maintains the fluctuation trajectory
regardless of the time series data; therefore,
Figure 1 Monthly Inbound Tourist Flows to China from January 1990 to December 2012.
Modeling the Fluctuations of Tourist Flows 945
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
the process aids in studying the complexity of
the time series data (An, Gao, Fang, Huang, &
Ding, 2014). The time series of inbound
tourist flows to China can be transformed
into a more understandable and limited sym-
bolic sequence using the coarse-graining
process. In this symbolic sequence, each
symbol denotes a distinct fluctuation pattern.
In the coarse-graining process, symbolic cat-
egories should be limited, and each symbol rep-
resents a distinct meta-pattern of fluctuation of
inbound tourist flows. The meta-pattern refers
to the distinct symbols in a symbolic sequence
(Li & Wang, 2006a). For inbound tourist
flows to China T(t), the fluctuation is
DT=TtTt1. The average monthly change
in inbound tourist flows to China from January
1990 to December 2012 is 487,690. We then
define four distinct symbols as follows:
Si=
R(DT487690),
r(0,DT,487690),
d(−487690 ,DT,0),
D(DT≤−487690),
(1)
where Ris a sharp-increase meta-pattern, ris a
small-increase meta-pattern, dis a small-
decrease meta-pattern, and Dis a sharp-
decrease meta-pattern. Here, R,r,d,andDare
measurements of the magnitude of the increase
and decrease of the fluctuations in inbound
tourist flows to China. Therefore, the time
series data of inbound tourist flows to China
are transformed into a symbolic sequence:
S={S1S2S3···},Si[(R,r,d,D).(2)
Network Construction
We can obtain various string combinations
using the distinctive symbols (R,r,d,D).
Each string of symbols denotes a distinct fluc-
tuation pattern of inbound tourist flows to
China. We define an n-string as a string
composed of nsymbols. For a given n, there
are a total of 4
n
different n-strings. In coarse-
graining processing, redundant information
increases as the number of symbol strings
increases (Li & Wang, 2006b). For this
reason, we defined three months (a quarter) as
a fluctuation pattern of inbound tourist flows
to China. The fluctuation patterns of inbound
tourist flows to China were 3-symbol strings
composed of R,r,d,andD. Therefore, the
fluctuation patterns and dynamics of inbound
tourist flows to China were investigated for
each six-month period. Because n¼3, 4
3
¼
64 3-strings, that is, (RRR,RRr,RRd,RRD,
RrR,RdR,RDR,... ), are theoretically poss-
ible. However, only 28 types of 3-strings actu-
ally appear. In symbol sequences of inbound
tourist flows to China, the 3-strings can rep-
resent different fluctuation patterns.
According to formulas (1) and (2), the
symbolic sequence of inbound tourist flows
to China is written in the form
{rDRrddrrdrdrdrrdddrrdrdrdrrrdd ... }. The
fluctuation patterns of inbound tourist flows
P
i
can be calculated by applying the following
formula:
Pi=(S3i2,S3i1,S3i).(3)
According to formula (3), the fluctuation pat-
terns of inbound tourist flows to China can be
expressed as {rDR,rdd,rrd,rdr,drr,ddd,rrd,
rdr,drr,rdd,... }. That is, the fluctuation pat-
terns evolve into each other with time {rDR
rdd rrd rdr drr ddd rrd
rdr drr rdd ... }. To identify the rule
of the transformation and detect significant
fluctuation patterns, we draw on complex
network theory to construct a weighted
network of inbound tourist flows to China.
The main idea of complex network theory is
946 Yongrui Guo et al.
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
to consider the relationships between various
parts of real complex systems as a complex
network (An et al., 2014). By analyzing the
structure of the network, we can better under-
stand the essential characteristics of the real
systems. In the symbolic sequence of fluctu-
ation patterns in inbound tourist flows to
China, the fluctuation patterns are defined as
nodes of the network, the transformations are
defined as edges, and the weight of an edge is
the number of transformations between the
two types of fluctuation patterns. The corre-
sponding network is shown in Figure 2.
Empirical Results
Important Fluctuation Patterns
We identified the significant fluctuation
patterns in the network of inbound tourist
flows using the degree and degree distribution
parameters of the complex network. The
concept of degree is the most fundamental
characteristic and measure of a node in a
network. The degree of a node in a complex
network is defined by the number of edges
directly connecting it to its neighbor. In this
paper, a node situated next to a given node is
considered its neighbor. The average degree
of a network is the average value of all node
degrees over the entire network. In undirected
networks, degree is a single number, but in
directed networks, nodes have two different
degrees, an in-degree and out-degree, corre-
sponding to the number of edges pointing
inward to and outward from those nodes. In
most cases, a node of higher degree is more
important than one of lower degree in a
network because it will have a more significant
influence on other nodes in the network in
terms of dynamics, information flows, and
data traffic, among other variables (Chen,
Wang, & Li, 2012). In this paper, the
Figure 2 The Weighted Network of Fluctuation Patterns in Inbound Tourist Flows to China.
Note: The greater the transition frequency, the thicker the line between nodes.
Modeling the Fluctuations of Tourist Flows 947
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
network of inbound tourist flows is a directed
network. The in-degree of the network of
inbound tourist flows represents other fluctu-
ation patterns transformed into a special
fluctuation pattern. The out-degree of the
network of inbound tourist flows represents
a special fluctuation pattern transformed into
other fluctuation patterns. Because the edge
is the sequential transformation between the
two nodes, the out-degree and in-degree of
each node are equal except for the first and
last nodes. In this paper, we use the out-
degree of the nodes to describe the degree
and degree distribution of the network of
inbound tourist flows. The degree distribution
of the network of inbound tourist flows can be
defined as follows:
p(k)=Ni
N,(4)
where N
i
denotes the number of nodes whose
degree equals kand Ndenotes the total
number of nodes in the network. A larger
out-degree of a special fluctuation pattern
implies a greater probability that this fluctu-
ation pattern will transform into another
pattern directly rather than through a series
of intermediate fluctuation patterns. A fluctu-
ation pattern of higher degree is more impor-
tant than one of lower degree.
We calculated the degree of every node in
the network of inbound tourist flows to
China. After that, we ranked the nodes in the
network from the highest to the lowest based
on the results degree calculation (Table 1).
According to Table 1, the first eight nodes
{rdr,RDr,RDd,RrD,rrd,DDR,drr,RDR}
have highest rank. The summed degree of
these nodes is 55.56%, and the degree of any
of these 8 nodes is more than 3% greater
than that of the others in the network of
inbound tourist flows, that is, 28.57% of the
nodes represent 55.56% of the degrees of the
network. This property means that these fluc-
tuation patterns have a significant role in
shaping macroscopic patterns of variations in
inbound tourist flows and in influencing
other patterns. In the process of transform-
ation of the fluctuation patterns, more trans-
formations should use these patterns as an
intermediate step. Identifying the out-degree
of nodes in the network of inbound tourist
flows contributes to understanding the fluctu-
ation rule and to forecasting future change.
For example, the fluctuation pattern RDd can
transform into three other types of patterns:
RrD,Rrd,andRdd. The transformation prob-
abilities are 0.57, 0.29, and 0.14, respectively.
The transformation probability between RDd
and RrD is greater than the other transform-
ation probabilities. The double-logarithmic
degree distribution and cumulative degree dis-
tribution plot of the network of inbound
tourist flows to China (Figure 3) shows few
nodes with a high degree value. Most of the
nodes are of lower degree value. Overall, the
degree distribution of the network of inbound
Table 1 Degree of Nodes in the Network of Inbound Tourist Flows to China
Node rdr RDr RDd RrD rrd DDR drr RDR ··· dDr
Degree 9 7 7 7 6 6 4 4 ··· 1
Rank 1 2 2 2 5 5 7 8 ··· 28
948 Yongrui Guo et al.
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
tourist flows to China follows a power-law dis-
tribution. Thus, only a handful of fluctuation
patterns significantly influence the fluctuations
in the inbound tourist flows to China.
Transformation Intermediary
In the transition process governing the fluctu-
ation patterns, we focus on the types of fluctu-
ation patterns that play an intermediary role in
the transformations. Thus, we can analyze the
betweenness centrality (BC) for each fluctu-
ation pattern in the network of inbound
tourist flows. In network theory, the between-
ness centrality of node vis the number of paths
from all nodes (except v) to all other nodes
that must pass through node v. The between-
ness centrality measures the intermediary or
middleperson power of a node. The between-
ness centrality of a node indicates its capability
to obtain and control resources or infor-
mation. Nodes with a high betweenness cen-
trality may have considerable influence
within a network by virtue of their control
over information or resources passing
between other nodes. The betweenness cen-
trality of a node vis given by the following
expression:
gv=
{i,j}
cij(v)
cij
,(5)
where g
v
is the betweenness centrality of a
node v,c
ij
is the total number of shortest
paths from node ito node j, and c
ij
(v) is the
number of those paths that pass through v.
The length of a path is the sum of the
weights of edges between iand j. The between-
ness centrality reveals the topological impor-
tance of nodes in its role in the transmission
of network information between each pair of
nodes. Therefore, the betweenness centrality
of a specific node can be explained as its
network influence. In the network of
inbound tourist flows, the network influence
of a specific node is the power of a fluctuation
pattern of inbound tourist flows to control or
affect other patterns in the network.
The betweenness centrality of the nodes in
the network of inbound tourist flows is
shown in Table 2. The differences among
nodes in betweenness centrality are evident.
The summed betweenness centrality of the
first 8 nodes is 63.18%, and the betweenness
centrality of any of these 8 nodes is more
Figure 3 Degree Distribution (a) and Cumulative Degree Distribution (b) in the Network of
Inbound Tourist Flows to China.
Modeling the Fluctuations of Tourist Flows 949
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
than 3% greater than that of the others; that
is, 28.57% nodes represent 63.18% of the
betweenness centrality of the network.
According to the statistics, six of the fluctu-
ation patterns {rdr,RDr,RDd,RrD,rrd,
DDR}listed in the top eight in degree rank
remain on the list in betweenness centrality.
This result means that these fluctuation pat-
terns are important intermediaries in the trans-
formation process for the fluctuation patterns
of inbound tourist flows. To a certain extent,
these fluctuation patterns can serve as a
precursor to the transformations between pat-
terns. These nodes are helpful in understand-
ing the inherent laws and transformation
information of the fluctuations in inbound
tourist flows to China. Furthermore, signifi-
cant differences among the betweenness cen-
trality of the nodes also suggest the presence
of higher volatility in the inbound tourist
flows to China.
Transformation Distance
Studying the shortest path length of the
network of inbound tourist flows can help us
understand the transition distance between
fluctuation patterns. The shortest path is the
minimum number of edges needed to connect
any two nodes. The average shortest path of
network Lis the average value of the shortest
path lengths of all of the connections between
two nodes, and it is defined as
L=2
N(N1)
ij
dij,(6)
where d
ij
is the distance between nodes iand j
and Nis the total number of nodes in the
network. As shown in Table 3, the shortest dis-
tance and the longest distance between nodes
are 1 and 8, respectively. The most frequentdis-
tances, 3 and 4, represent more than 54% of the
cases in the network of inbound tourist flows.
The calculated value of the average shortest
path is 3.38. Therefore, if one type of fluctuation
pattern transforms into another, it will basically
change via three or four types of patterns.
Different types of fluctuation patterns rarely
Table 2 Betweenness Centrality of Nodes in the Network of Inbound Tourist Flows to China
Vertex rdr RDr RrD DDR DdR RDd rrd rDr ··· DdD
BC/% 14.71 11.53 7.46 7.45 6.81 6.75 4.65 3.78 ··· 0.32
Rank 1 2 3 4 5 6 7 8 ··· 28
Table 3 Frequencies of Shortest Path
Distances in the Network of Inbound Tourist
Flows to China
Distance Frequency Proportion (%)
1 60 7.9
2 138 18.3
3 213 28.2
4 199 26.3
5 100 13.2
6 36 4.8
7 9 1.2
8 1 0.1
950 Yongrui Guo et al.
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
transform into each other directly (distance of 1,
which accounts for 7.9% of cases). In the longest
path of transformation, a type of fluctuation
pattern transforms via 8 types of patterns, but
this case is infrequent (0.1% of cases). A trans-
formation occurred approximately every three
to four quarters. This information is used as a
basis for predicting changes in inbound tourist
flows in the next year. The results of the analysis
demonstrate that the transformations between
fluctuation patterns of inbound tourist flows
occur frequently.
Conclusions
This paper analyzed the fluctuation patterns of
monthly inbound tourist flows to China from
January 1990 to December 2012 using a
complex network approach. We constructed
a weighted network of inbound tourist flows
to China. The nodes represent 28 fluctuation
patterns of inbound tourist flows to China,
the edges are the transformations between
nodes, and the weight of an edge is the
number of transformations between the two
types of fluctuation patterns.
The most important nodes in the network are
rdr,RDr,RDd,RrD,rrd, and DDR. The
degree and betweenness of these nodes are
greater than those of other nodes in the
network of inbound tourist flows. The
summed degree and betweenness of these
nodes reach values of 46.67% and 52.58%,
respectively. In the process of transformation
of fluctuation patterns, more transformations
should use these patterns as an intermediate
step. We can identify significant fluctuation
patterns of inbound tourist flows using these
topologically important nodes in the network.
These significant fluctuation patterns of
inbound tourist flows play a key role in
pattern transformation and can be viewed as
the prelude to changes in inbound tourist
flows. These fluctuation patterns are helpful
in understanding the inherent laws and trans-
formation information related to the fluctu-
ation in inbound tourist flows to China. The
average transition distance was 3.38, and a
transformation occurred approximately every
34 quarters. These results demonstrate that
the transformations between fluctuation pat-
terns of inbound tourist flows occur frequently.
This paper analyzed the complex character-
istics of the fluctuation patterns of monthly
inbound tourist flows to China from the perspec-
tive of network topology. This method was of
guiding significance in identifying the important
fluctuation patterns and understanding the
inherent laws of the fluctuations in tourist
flows. The statistical properties of fluctuations
in tourist flows are important for modeling the
complex dynamics of tourist flows and are of
great significance for practical applications such
as tourist flow risk estimation and tourism flow
forecasting. According to this method and the
results in this paper, a forecasting model can be
built using the transformation intermediary, the
transformation probability and the transform-
ation time. This model differs from previous
tourism demand forecasting models because the
model is based on the fluctuation patterns of
tourist flows but not on the time series itself.
This difference is a direction for future research.
Acknowledgement
This work was supported by the National
Natural Science Foundation of China (Grant
No. 41171121, 41301134).
References
An, H., Gao, X., Fang, W., Huang, X., & Ding, Y. (2014).
The role of fluctuating modes of autocorrelation in
Modeling the Fluctuations of Tourist Flows 951
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
crude oil prices. Physica A: Statistical Mechanics and its
Applications,393, 382– 390.
Assaf, A. G., Barros, C. P., & Gil-Alana, L. A. (2011).
Persistence in the short- and long-term tourist arrivals to
Australia. Journal of Travel Research,50(2), 213– 229.
Athanasopoulos, G., & de Silva, A. (2012). Multivariate
exponential smoothing for forecasting tourist arrivals.
Journal of Travel Research,51(5), 640– 652.
Bonanno, G., Caldarelli, G., Lillo, F., Micciche
´, S., Vande-
walle, N., & Mantegna, R. N. (2004). Networks of equi-
ties in financial markets. The European Physical Journal
B Condensed Matter and Complex Systems,38(2),
363– 371.
Chan, F., Lim, C., & McAleer, M. (2005). Modelling
multivariate international tourism demand and vola-
tility. Tourism Management,26(3), 459– 471.
Chang, Y. W., & Liao, M. Y. (2010). A seasonal ARIMA
model of tourism forecasting: The case of Taiwan. Asia
Pacific Journal of Tourism Research,15(2), 215– 221.
Chen, G. R., Wang, X. F., & Li, X. (2012). Introduction to
complex networks: Models, structures and dynamics.
Beijing: Higher Education Press.
Cho, V. (2003). A comparison of three different
approaches to tourist arrival forecasting. Tourism Man-
agement,24(3), 323– 330.
Cho, V. (2009). A study on the temporal dynamics of
tourism demand in the Asia Pacific region. International
Journal of Tourism Research,11(5), 465– 485.
Chu, H. P., Yeh, M. L., & Chang, T. Y. (2014). Are
visitor arrivals to China stationary? An empirical note.
Asia Pacific Journal of Tourism Research,19(2), 248
256.
CNTA. (2013). The yearbook of China tourism statistics
2013. Beijing: China Tourism Press.
Coshall, J. T. (2006). Time series analyses of UK outbound
travel by air.Journal of Travel Research,44(3), 335– 347.
Cuccia, T., & Rizzo, I. (2011). Tourism seasonality in cul-
tural destinations: Empirical evidence from Sicily.
Tourism Management,32(3), 589– 595.
De Cantis, S., Ferrante, M., & Vaccina, F. (2011). Seaso-
nal pattern and amplitude – a logical framework to
analyse seasonality in tourism: An application to bed
occupancy in Sicilian hotels. Tourism Economics,
17(3), 655– 675.
Fourie, J., & Santana-Gallego, M. (2011). The impact of
mega-sport events on tourist arrivals. Tourism Manage-
ment,32(6), 1364– 1370.
Gil-Alana, L. A. (2005). Modelling international monthly
arrivals using seasonal univariate long-memory pro-
cesses. Tourism Management,26(6), 867– 878.
Goh, C., & Law, R. (2002). Modeling and forecasting
tourism demand for arrivals with stochastic nonstation-
ary seasonality and intervention. Tourism Management,
23(5), 499– 510.
Gursoy, D., Parroco, A. M., & Scuderi, R. (2013). An
examination of tourist arrivals dynamics using short-
term time series data: A space-time cluster approach.
Tourism Economics,19(4), 761– 777.
Gustavsson, P., & Nordstrom, J. (2001). The impact of
seasonal unit roots and vector ARMA modelling on
forecasting monthly tourism flows. Tourism Econ-
omics,7(2), 117– 133.
Hadavandi, E., Ghanbari, A., Shahanaghi, K., & Abba-
sian-Naghneh, S. (2011). Tourist arrival forecasting by
evolutionary fuzzy systems. Tourism Management,
32(5), 1196– 1203.
Higham, J., & Hinch, T. (2002). Tourism, sport and
seasons: The challenges and potential of overcoming
seasonality in the sport and tourism sectors. Tourism
Management,23(2), 175– 185.
Jang, S. C. (2004). Mitigating tourism seasonality: A
quantitative approach. Annals of Tourism Research,
31(4), 819– 836.
Kim, J. H., & Moosa, I. A. (2001). Seasonal behaviour of
monthly international tourist flows: Specification and
implications for forecasting models. Tourism Econ-
omics,7(4), 381– 396.
Kuo, H., Liu, K. E., & Chen, C. (2014). Modeling Japa-
nese tourism demand for Asian destinations: A
dynamic AIDS approach. Asia Pacific Journal of
Tourism Research,19(1), 86– 102.
Lacasa, L., Luque, B., Ballesteros, F., Luque, J., & Nuno,
J. C. (2008). From time series to complex networks: The
visibility graph. Proceedings of the National Academy
of Sciences of the United States of America,105(13),
4972– 4975.
Lean,H.H.,&Smyth,R.(2009).Asianfinancialcrisis,
avian flu and terrorist threats: Are shocks to Malay-
sian tourist arrivals permanent or transitory? Asia
Pacific Journal of Tourism Research,14(3),
301–321.
Li, G., Song, H., & Witt, S. F. (2005). Recent develop-
ments in econometric modeling and forecasting.
Journal of Travel Research,44(1), 82– 99.
Li, G., Wong, K. K., Song, H., & Witt, S. F. (2006).
Tourism demand forecasting: A time varying parameter
error correction model. Journal of Travel Research,
45(2), 175– 185.
Li, P., & Wang, B. H. (2006a). An approach to Hang Seng
Index in Hong Kong stock market based on network
952 Yongrui Guo et al.
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
topological statistics. Chinese Science Bulletin,51(5),
624–629.
Li, P., & Wang, B. H. (2006b). A dynamic model of Hang
Seng Index based on complex network eigenvectors.
Systems Engineering,24(3), 73– 77.
Li, P., & Wang, B. H. (2007). Extracting hidden fluctu-
ation patterns of Hang Seng stock index from network
topologies. Physica A: Statistical Mechanics and its
Applications,378(2), 519– 526.
Lim, C., & McAleer, M. (2001). Monthly seasonal vari-
ations – Asian tourism to Australia. Annals of
Tourism Research,28(1), 68– 82.
Nadal, J. R., Font, A. R., & Rossello, A. S. (2004). The
economic determinants of seasonal patterns. Annals of
Tourism Research,31(3), 697– 711.
Shareef, R., & McAleer, M. (2005). Modelling inter-
national tourism demand and volatility in small island
tourism economies. International Journal of Tourism
Research,7(6), 313– 333.
Song, H., & Li, G. (2008). Tourism demand modelling
and forecasting a review of recent research. Tourism
Management,29(2), 203– 220.
Song, H., Li, G., Witt, S. F., & Athanasopoulos, G.
(2011). Forecasting tourist arrivals using time-varying
parameter structural time series models. International
Journal of Forecasting,27(3), 855– 869.
Song, H., Lin, S. S., Zhang, X. Y., & Gao, Z. X. (2010).
Global financial/economic crisis and tourist arrival
forecasts for Hong Kong. Asia Pacific Journal of
Tourism Research,15(2), 223– 242.
Song, H., & Witt, S. F. (2006). Forecasting international
tourist flows to Macau. Tourism Management,27(2),
214–224.
Tao, W., & Ni, M. (2010). Study on the comparison of
tourism demand forecast between China and western
countries: Basic theory and models. Tourism Tribune,
25(8), 12– 17.
Turner, L. W., & Witt, S. F. (2001). Forecasting
tourism using univariate and multivariate structural
time series models. Tourism Economics,7(2), 135–
147.
Vergori, A. S. (2012). Forecasting tourism demand:
The role of seasonality. Tourism Economics,18(5),
915–930.
Wong, K. K., Song, H., Witt, S. F., & Wu, D. C. (2007).
Tourism forecasting: To combine or not to combine?
Tourism Management,28(4), 1068– 1078.
Yan, M., & Wall, G. (2003). Disaggregating visitor flows:
The example of China. Tourism Analysis,7(3–4), 191 –
205.
Yang, Y., & Wong, K. K. F. (2013). Spatial distribution of
tourist flows to China’s cities. Tourism Geographies,
15(2), 338– 363.
Yang, Y., Pan, B., & Song, H. (2014). Predicting hotel
demand using destination marketing organization’s
web traffic data. Journal of Travel Research,53(4),
433–447.
Zhang, J. H., Zhou, H. X., Jiang, L., & Wang, Y. G.
(2010). Network topologies of Shanghai stock index.
Physics Procedia,3(5), 1733– 1740.
Modeling the Fluctuations of Tourist Flows 953
Downloaded by [Stephen F Austin State University] at 04:23 05 August 2015
... This paper focused on the spatial structure and pattern of tourism flows according to Stewart (1996) [26], Mckercher (2008) [27], and Guo (2014) [28], a complex network approach able to study well the spatial structure and pattern of tourism flows, thus using the method of social network analysis. This study considering the advantages and disadvantages of traditional data and online big data, adopted the social network analysis (SNA) method and applied search query volume data provided by Baidu to investigate the characteristics of the network spatial structure of the urban tourism information flows in China. ...
... Based on the relevant theories and methods of traditional research [5,28], this paper makes full use of the advantages of big data in tourism research [6,7,34,35], and studies the information flow network of China's tourism through the Baidu index. Compared with existing research, this paper provides a perspective for the study of large-scale tourism flows using relatively easy data sources. ...
Article
Full-text available
The characteristics of information flow, as represented by the Baidu index, reflect the pattern of tourism flows between different cities. This paper is based on the Baidu index and applies the seasonal concentration index and social network analysis (SNA) methods to study the spatial structure characteristics of tourism flows in China. The results reveal that: (1) both the search volume of the Baidu index in different cities and the online attention to different scenic areas exhibit obvious spatial heterogeneity and seasonal differences; (2) regions with strong tourism flow connections mainly occur in the areas between metropolises or among the inner cities of urban agglomerations, which are largely distributed on the southeast side of the Heihe–Tengchong Line; (3) the development of the whole tourism flow network in China is low, with an unbalanced development between tourism supply and demand, indicating that tourism resources are concentrated in a few cities and that most of the information interaction among cities occurs in core areas, while a weak interaction is observed in peripheral areas; (4) cities like Beijing and Wuhan attain obvious advantages in regard to their tourism resources, whereas other cities, including Beijing, Shanghai, Shenzhen and Guangzhou, exhibit a high demand for tourism. Moreover, tourism information flow networks are concentrated in several cities with an important role in the Chinese urban system, such as Beijing, Wuhan, and Chengdu, because they contain abundant tourism resources, well-developed transportation systems and advanced economic and societal development levels. (5) Cities such as Beijing, Lhasa, Wuhan, and Zhengzhou possess numerous advantages due to structural holes, and they thus occur at an advantageous position in the tourism flow network.
... Using FDA to analyze real-time tourist volume data from 56 major attractions in Beijing, we identify potential factors influencing changes in tourist volume and uncover subtle patterns within these fluctuations. As highlighted in prior research, understanding the statistical characteristics of tourist volume fluctuations has significant practical value for constructing comprehensive tourist volume dynamic models and assessing and predicting tourist volume risks (Guo et al., 2015;Kádár & Gede, 2021). ...
Article
Understanding the spatiotemporal dynamics of tourist volume is crucial for effective tourism management and planning. However, existing tourism data analysis methods often fail to capture the complex, continuous fluctuations and temporal variations in tourist behavior. To address this challenge, we apply functional data analysis (FDA) in the tourism industry to provide a more nuanced understanding of tourist volume dynamics. Specifically, we perform FDA on real-time tourist volume data from 56 major attractions in Beijing, China, revealing intrinsic fluctuation patterns, key factors driving tourist arrivals, and the dynamic characteristics of attractions across temporal scales. Our findings enhance the ability to optimize attraction management, marketing strategies, and policy-making, while also advancing tourism data science by integrating FDA. This approach fills the methodological gap and offers a comprehensive framework for exploring the spatiotemporal complexities of tourism data.
... The network analysis approach is used to analyze the dynamics of the networks corresponding to time series with yearly data. The evolution of links is analyzed using the natural or horizontal visibility algorithm (Baggio, 2014a;Baggio & Sainaghi, 2016;Chung et al., 2020;Guo et al., 2015;Sainaghi & Baggio, 2017). Stpartial patterns of tourism networks. ...
Article
This study analyzes the tourism network and destinations after the COVID-19 pandemic using social network analysis (SNA). Analysis of 789 destinations in Thailand has found that the destinations are connected by 1,1175 tourism routes. The network is a sparse network with a low network density. It seems to have a scale-free property that reflects that most destinations have low connectivity and a small number of destinations have high connectivity. The network has a large average path length and low clustering coefficient. Different roles of destinations are identified based on degree, betweenness and closeness centrality. The findings draw implications for vitalizing the sector.
... In China, tourism has developed rapidly since the country's reform and opening up in 1978. In 1998, tourism was selected as a new growth pole of the national economy, and in 2009 it became a strategic pillar industry [46]. In recent years, China has had the fastest growth rate of tourism development in the world [47]. ...
Article
Full-text available
Existing research has noted a clear interaction between touristification and commercial gentrification; however, the differences between these two coexisting but distinct phenomena require further research. This study uses online big data and quantitative methods to explore the relationship between touristification and commercial gentrification. Taking Yuzhong District in Chongqing as an example, this study constructs an inter-attraction network based on 1306 itineraries extracted from online travel diaries, develops a method to evaluate community tourism centrality based on network analysis, and examines the correlation between community tourism centrality, touristification, and commercial gentrification. The results suggest that attractions with historical value, unique local landscapes, and mixed functions show greater tourism centrality in the tourist flow network. Attractions with similar themes are more likely to be included in one travel route, and the influence of distance is insignificant at the district level. Communities with higher tourism centrality are clustered in old city areas with a rich historic heritage and have experienced profound commercialisation. Although similar, touristification is primarily a bottom-up process, while commercial gentrification tends to be more involved with the top-down urban planning process. This study contributes to the methodological development of network analysis in tourism research and advances the understanding of the different mechanisms of touristification and commercial gentrification.
... The issue of overcrowding at historically significant tourist sites and the measurement of their capacity is addressed in the studies conducted by G. Liberatore, P. Biagioni, C. Ciappei, and C. Francini (2022). The article by Guo, Yo., Zhang, J., Yang, Ya., and Zhang, H. (2015) presented the results of a study on models of fluctuations and dynamics of inbound tourist flows. Their research aimed to gain insights into the natural laws and transformations that govern the fluctuations of tourist flows. ...
Article
Full-text available
The aim of the study is to define methodological approaches and methods of statistical analysis of arrivals to the centers of excursion tourism. In destinations with highly appealing attractions or objects, accurately accounting for the number of tourists visiting them is crucial. In Chernivtsi, such an attraction or object is the former Residence of the Bukovinian Metropolitans, a UNESCO monument. The statistical analysis traces the dynamics of the number of excursion visitors to the former Residence. It identifies the factors that influenced it, including the inclusion of the site in the UNESCO list, Russian aggression since 2014, and the COVID‑19 pandemic. Based on the statistical reporting data analysis of the Historical and Architectural Museum Complex of Chernivtsi National University, three periods of excursion activity from 2000–2021 were identified. The first period (2000–2017) is characterized by rapid growth due to the inclusion of the former Residence in UNESCO. The second period (2017–2019) is a period of stagnation of excursion activity with minor fluctuations. The hypothesis and data analysis confirmed the thesis that the occupation of Crimea by Russia had an impact on the redistribution of tourist flows, particularly during the May holidays, within Ukraine, specifically in Chernivtsi. The third period (2019–2020), characterized by a sharp decline followed by the same dynamic growth (2020–2021), was caused by a force majeure event of global proportions – the COVID‑19 pandemic.
... In recent years, some scholars used the above theory to study tourism, mainly involving tourism flow and the size and spatial distribution of regional tourism places [1][2][3][15][16][17][18][19][20][21][40][41][42][43]. ...
Article
Full-text available
It is well known that Zipf’s rank-size law is powerful to investigate the rank-size distribution of tourist flow. Recently, widespread attention has been drawn to investigating the impacts of COVID-19 on tourism for its sustainability. However, little is known about the impacts of COVID-19 on the rank-size distribution of regional tourism central places. Taking Guangdong-Hong Kong-Macao Greater Bay Area as a research case, this article aims to examine the fractal characteristics of the rank-size distribution of regional tourism central places, revealing the impacts which COVID-19 has on the rank-size distribution of regional tourism central places. Based on the census data over the years from 2008 to 2021, this paper reveals that before COVID-19, the rank-size distribution of the tourism central places in Guangdong-Hong Kong-Macao Greater Bay Area appears monofractal, and the difference in the size of the tourism central places has a tendency to gradually decrease; in 2020, with the outbreak of COVID-19, the characteristic of the rank-size distribution shows that the original monofractal is broken into multifractal; in 2021, with COVID-19 becoming under control, the structure of tourism size distribution, changes into bifractal based on the original multifractal, showing that the rank-size distribution of tourism central places in Guangdong-Hong Kong-Macao Greater Bay Area becomes more ideal and the tourism order becomes better than the last year. The results obtained not only fill in the gap about the impacts of COVID-19 on tourism size distribution, but also contribute to the application of fractal theory to tourism size distribution. In addition, we propose some suggestions to the local governments and tourism authorities which have practical significance to tourism planning.
... For example, Shih [18] applied this methodology to analyze drive tourism destinations focusing on node ties, while Leung, Wang [14] investigated movement patterns of overseas tourists in Beijing during the Olympics. By utilizing complex network, Guo, Zhang [19] examined the fluctuation patterns of monthly inbound tourist flows in China sharing valuable insights on the nature of tourism demand. Shao, Huang [10] applied this approach to illustrate the evolution of international tourist flows over 1995-2018 focusing, in particular, on properties of tourist flows network as well as the roles and functions of countries/regions within it. ...
Article
Full-text available
This study applies complex network analysis to examine global tourist flows network in the context of Belt and Road Initiative (BRI). Using tourist flows data between 221 countries/regions over 1995–2018, we investigate the nature and development patterns of structural properties of global network as well as factors influencing its formation. The descriptive analysis indicates that global tourist network was a sparse network with small world network characteristics. According to centrality characteristics, China showed the most influence in the BRI group, while Germany and the United States possessed key roles among non-BRI countries/regions. Exploratory analysis demonstrated significant influence of gravity variables in global, BRI and non-BRI tourist networks. This research advances existing tourism theory and provides practical implications for policymakers.
... 合作网络的基本结构与特征 网络核心-边缘结构分析,核心区合作关系较强,边缘区合作 关系较弱 [20][21] 同一凝聚子群内的节点合作关系较强,聚类指数越高的节点间 合作关系越强 [22][23] 具有稳定二方、三方关系的节点,合作强度较大,合作关系较 稳定 节点中心度用于判别节点的交通和旅游功能 [24][25][26] ,景点核密度用 于验证节点的功能,并结合二方、三方关系总结旅游地合作模式 [18,21] ,但截断值的选取对于旅游流网络的整体结构有重要 影响,甚至常常导致运算结果出现差异。目前,学界关于截断值的选取尚未形成统一的 标准,常见的取值方法包含差值法、观察法、实验法等,截断值的取值差异较大 [18,21,27] 合作模式二: "旅游核心+交通核心" ,如: "黄山市+合肥市" "南平市+福州市" 。高 铁开通后,交通区位优势开始转移,新兴高铁城市,尤其是高铁枢纽城市的交通区位条 件大幅提升,其与高铁沿线旅游核心之间的合作网络不断加强。合福高铁开通前,杭州 市是上海、苏州、无锡等长三角城市进入黄山市的交通门户,长期以来与黄山市保持着 稳定的旅游合作关系 [14] 。合福高铁开通后,杭州只能通过上饶等城市中转进入黄山市, 其在合作网络中的交通地位大幅下降,而合肥市作为我国"米"字型高铁的衔接处,成 表 7 网络节点核心功能识别 Table 7 Core function recognition of network nodes 1) After the opening of the Hefei-Fuzhou high-speed railway, the tourists' cognition degree, the willingness to travel, the willingness to revisit the tourist destinations along the route have increased. The cooperative relation between different tourism destinations has been strengthened, and the cooperative network has formed a remarkable core-edge structure. ...
Article
Purpose This paper gathers tourism digital footprint from online travel platforms, choosing social network analysis method to learn the structure of destination networks and to probe into the features of tourist flow network structure and flow characteristics in Guilin of China. Design/methodology/approach The digital footprint of tourists can be applied to study the behaviors and laws of digital footprint. This research contributes to improving the understanding of demand-driven network relationships among tourist attractions in a destination. Findings (1) Yulong River, Yangshuo West Street, Longji Terraced Fields, Silver Rock and Four Lakes are the divergent and agglomerative centers of tourist flow, which are the top tourist attractions for transiting tourists. (2) The core-periphery structure of the network is clearly stratified. More specifically, the core nodes in the network are prominent and the core area of the network has weak interaction with the peripheral area. (3) There are eight cohesive subgroups in the network structure, which contains certain differences in the radiation effects. Originality/value This research aims at exploring the spatial network structure characteristics of tourism flows in Guilin by analyzing the online footprints of tourists. It takes a good try to analyze the application of network footprint with the research of tourism flow characteristics, and also provides a theoretical reference for the design of tourist routes and the cooperative marketing among various attractions.
Article
Full-text available
The applications of social network analysis to the world tourism network are scarce, and a research update is long overdue. The goal of this research is to examine the topology of the world tourism network and to discuss the meaning of its characteristics in light of the current situation. The data used for the analysis comprise 193 target countries, 242 source countries, and 17,022 links, which is an overall 1,448,285,894 travels in 2018. Social network analysis is applied to the data to determine network topological and diffusion properties, as well as the network structure and its regularities (does it behave more as a social or a technological/biological network?). While results presented in this paper give a thorough insight into the world tourism network in the year 2018, they are only a glimpse in comparison to the possibilities for further research.
Article
Full-text available
In this work we present a simple and fast computational method, the visibility algorithm, that converts a time series into a graph. The constructed graph inherits several properties of the series in its structure. Thereby, periodic series convert into regular graphs, and random series do so into random graphs. Moreover, fractal series convert into scale-free networks, enhancing the fact that power law degree distributions are related to fractality, something highly discussed recently. Some remarkable examples and analytical tools are outlined to test the method’s reliability. Many different mea-sures, recently developed in the complex network theory, could by means of this new approach characterize time series from a new point of view. From time series to complex networks: The visibility graph. Available from: https://www.researchgate.net/publication/301232812_From_time_series_to_complex_networks_The_visibility_graph [accessed Jun 29, 2017].
Article
Full-text available
This study uses the web traffic volume data of a destination marketing organization (DMO) to predict hotel demand for the destination. The results show a significant improvement in the error reduction of ARMAX models, compared with their ARMA counterparts, for short-run forecasts of room nights sold by incorporating web traffic data as an explanatory variable. These empirical results demonstrate the significant value of website traffic data in predicting demand for hotel rooms at a destination, and potentially even local businesses' future revenue and performance. The implications for future research on using big data for forecasting hotel demand is also discussed.
Book
Complex networks such as the Internet, WWW, transportation networks, power grids, biological neural networks, and scientific cooperation networks of all kinds provide challenges for future technological development. The first systematic presentation of dynamical evolving networks, with many up-to-date applications and homework projects to enhance study. The authors are all very active and well-known in the rapidly evolving field of complex networks. Complex networks are becoming an increasingly important area of research. Presented in a logical, constructive style, from basic through to complex, examining algorithms, through to construct networks and research challenges of the future.
Article
Half-Title Page Title Page Copyright Page Table of Contents About the Authors Preface Acknowledgements
Article
Most tourism destinations are affected by seasonality. Seasonal demand causes various problems for local firms and administrations hampering the efficient use of available facilities and the development of local capabilities. This paper discusses, from an econometric point of view, another important issue stemming from strong seasonality: the effect on forecasting tourist flows. This aspect of seasonality is addressed through an analysis of tourist arrivals in the Province of Lecce, southern Italy.
Article
The purpose of this study is to examine the development of Italian tourist areas (circoscrizioni turistiche) through a cluster analysis of short time series. The technique is an adaptation of the functional data analysis approach developed by Abraham et al (2003), which combines spline interpolation with k-means clustering. The findings indicate the presence of two patterns (increasing and stable) averagely characterizing groups of territories. Moreover, tests of spatial contiguity suggest the presence of 'space–time clusters'; that is, areas in the same 'time cluster' are also spatially contiguous. These findings appear to be more robust in particular for those series characterized by an increasing trend.
Article
This paper investigates the expenditure allocation of Japanese international tourism in its five major Asian destinations, China, Hong Kong, Korea, Taiwan, and Thailand. The dynamic of linear approximation the almost ideal demand system is used to determine the long-run equilibrium while the short-run dynamics are represented by an error correction mechanism. The empirical results indicate that the changes in market shares of Japanese outbound tourism are significantly influenced by the changes in tourists' expenditure, rather than the changes in relative tourism prices. The results show that Japan expenditure rises, the market share of Taiwan and Thailand declines, while Korea benefits. In addition, price competitiveness is important for Japanese demand for Korea, but is relatively unimportant for the other destinations.
Article
The tourism industry is one of the earliest industries in China to be opened up to the world, with a more open market and relatively sufficient competition. In this article, we use the panel seemingly unrelated regressions augmented Dickey–Fuller (SURADF) tests advanced by Breuer, McNown, and Wallace [(2001). Misleading inferences from panel unit-root tests with an illustration from purchasing power parity. Review of International Economics, 9(3), 482–493] to investigate whether visitor arrivals to China are stationary for the period 1995–2010. The empirical results from numerous conventional unit root tests indicate that tourist arrivals from 18 countries studied are non-stationary; however, when Breuer et al.'s (2001) panel SURADF tests are conducted, evidence of a unit root in visitor arrivals is found in only 13 countries. From these results, one particularly important policy implication for China emerges.
Article
Autocorrelation exists in the crude oil price due to price inertia, the cobweb theorem, model errors, etc. Many researchers have studied the fluctuation of the crude oil price, but few have focused on the autocorrelation fluctuation in crude oil prices. Exploring the fluctuating rules of autocorrelation can aid in understanding the fluctuating mechanism of crude oil prices. To study the role of fluctuating modes of autocorrelation in crude oil prices, which have time series characteristics, this study selected international crude oil spot prices as sample data to employ the methods of statistical physics. The fluctuating modes of autocorrelation were defined by the autocorrelation coefficient, symbolization, and a coarse-graining process. We set the modes as nodes and the transformation between modes as edges; the fluctuating mode weight network of autocorrelation was then built. Thus, the study of autocorrelation fluctuation was transformed to a network study. Then, certain aspects, such as the statistical properties, the "small-world" behavior, and the transmission medium in the network, could be analyzed using complex network theory and analytical methods. The periodicity of the fluctuation was calculated using a spectral analysis method. This study not only describes the fluctuation of the time series more precisely than other methods but also provides ideas for methods of studying the fluctuation of univariate autocorrelations.