Tourism Demand Forecasting Research Articles

A Forecast Model of Tourism Demand Driven by Social Network Data

To improve the forecasting accuracy of tourism demand through forecasting model and data sources, this paper takes the social network data as an entry point, and collects the social network data by the web crawler, then quantifies the data based on the sentiment analysis of the BERT model. This paper uses structured variables such as social network data, weather, holidays, etc. to build a tourism demand forecasting model based on Gradient Boosting Regression Trees. At last, take Huang Shan as example, use actual statistics of passenger terminal and social network data to do an empirical analysis of Huang Shan tourism demand forecasting. Compared with the existing model and introduce ablation study to verify the effectiveness of the considered factors. The result shows that the model based on social network data has improved the forecasting accuracy from the existing ones, ablation study shows social network data helps to improve the accuracy of tourism demand forecasting.

Forecasting the number of incoming tourists using Arima model: case study from Armenia

This paper summarizes the arguments and counterarguments within the scientific discussion on the issue of forecasting tourism demand and touristic flows. During COVID-19 tourism sphere suffered a lot in the whole world. Many countries try to do forecasts and make recovery plans for tourism. Tourism has been a growing sphere in Armenia in recent years. However, the number of incoming tourists decreased by 80 percent because of the pandemic. The main purpose of the research is to forecast tourism demand in the Republic of Armenia. Systematization of scientific sources and approaches for solving the problem identified many methods and models for doing forecasts. The variables used to depend on the method selected. For gaining the research goal, the study was carried out in the following logical sequence: 1) discussion on some literature sources; 2) analysis of the current situation of tourism in Armenia; 3) interpretation of forecast results; 4) providing some recommendations. The methodological tool of the research was mainly the ARIMA method. The data rest on the publications of the Statistical Committee of the Republic of Armenia. Time series for the number of incoming tourists include from 2001-Q1 till 2019-Q4 data. 2020 was not included in the model, as there was a sharp decline. Besides, in the second quarter of 2020, there were no tourists at all because of restrictions and flight cancellations. The obtained data show that if there were no pandemic, the number of incoming tourists would increase on average by 12.81% in 2021, 13.42% – in 2022, and 13.66% – in 2023. The results are realistic. The tourism sphere is expected to grow in 2021. This paper suggested some steps for recovering and restoring tourism, particularly by using aggressive marketing strategies, word-of-mouth, influencer marketing, etc. The research results could be useful for state organs of the sphere to forecast their strategic policies. The applied approach and suggestions may be helpful in many countries which try to restart tourism after the pandemic.

(Предвидување на динамичката побарувачка во туризмот со користење на вештачки невронски мрежи)

Planning tourism development means preparing the destination for coping with uncertainties as tourism is sensitive to many changes. This study tested two types of artificial neural networks in modeling international tourist arrivals recorded in Ohrid (North Macedonia) during 2010–2019. It argues that the MultiLayer Perceptron (MLP) network is more accurate than the Nonlinear AutoRegressive eXogenous (NARX) model when forecasting tourism demand. The research reveals that the bigger the number of neurons may not necessarily lead to further perfor- mance improvement of the model. The MLP network for its better performance in modeling series with unexpected challenges is highly recommended for forecasting dynamic tourism demand

Pronóstico de la demanda turística de Chile basado en modelos lineales y no lineales estacionales

TIn this paper, the Chilean tourism demand experienced during the period 2000‑2018 has been modelled using a linear regression model with dichotomous variables (MRL) and an ARIMA model with a seasonal component (SARIMA). The results show that the SARIMA approach is more effective in replicating the non‑stationary, non‑linear behaviour and the presence of seasonality in the series, with the forecasts obtained from this model presenting a low rate of error, in this case ‑5.6% for outbound tourism and ‑5.9% for inbound tourism. In consequence, this approach may be an effective tool for forecasting forecast tourism demand in the short term, and support for planning and management in the sector in the face of fluctuations in tourism demand.

Modelling and Forecasting Inbound Tourism Demand to Croatia using Artificial Neural Networks: A Comparative Study

Tourism demand is the basis on which all commercial decisions concerning tourism ultimately depend. Accurate estimation of tourism demand is essential for the tourism industry because it can help reduce risk and uncertainty as well as effectively provide basic information for better tourism planning. The purpose of this study is to develop the optimal forecasting model that yields the highest accuracy when compared to the forecast performances of three different methods, namely Artificial Neural Network (ANN), Exponential Smoothing, and Box-Jenkins methods for forecasting monthly inbound tourist flows to Croatia. Prior studies have been applied to forecast tourism demand to Croatia based on time series models and casual methods. However, the monthly and comparative tourism demand forecasting studies using ANNs are still limited, and this paper aims to fill this gap. The number of monthly foreign tourist arrivals to Croatia covers the period between January 2005-December 2019 data were used to build optimal forecasting models. Forecasting performances of the models were measured by Mean Absolute Percentage Error (MAPE) statistics. As a result of the experiments carried out, when compared to the forecasting performances of various models, 12 lagged ANN models, which have [4-3-1] architecture, were seen to perform best among all models applied in this study. Considering both the empirical findings obtained from this study and previous studies on tourism forecasting, it can be seen that ANN models that do not have any negativities (such as over-training, faulty architecture, etc.) produce successful forecasting results when compared with results generated by conventional statistical methods.

Forecasting Foreign Visitors using SARIMAX Models with the Exogenous Variable of Demand Decrease

In this study, we consider the problem of forecasting the number of inbound foreigners visiting Korea. Forecasting tourism demand is an essential decision to plan related facilities and staffs, thus many studies have been carried out, mainly focusing on the number of inbound or outbound tourists. In order to forecast tourism demand, we use a seasonal ARIMA (SARIMA) model, as well as a SARIMAX model which additionally comprises an exogenous variable affecting the dependent variable, i.e., tourism demand. For constructing the forecasting model, we use a search procedure that can be used to determine the values of the orders of the SARIMA and SARIMAX. For the exogenous variable, we introduce factors that could cause the tourism demand reduction, such as the 9/11 attack, the SARS and MERS epidemic, and the deployment of THAAD. In this study, we propose a procedure, called Measuring Impact on Demand (MID), where the impact of each factor on tourism demand is measured and the value of the exogenous variable corresponding to the factor is determined based on the measurement. To show the performance of the proposed forecasting method, an empirical analysis was conducted where the monthly number of foreign visitors in 2019 were forecasted. It was shown that the proposed method can find more accurate forecasts than other benchmarks in terms of the mean absolute percentage error (MAPE).

Modeling and Forecasting of Tourism Demand in Malaysia

The pattern of foreign tourist demand to Malaysia is analysed and forecasted using time series method and non-linear technique. There are nine selected countries that contribute a lot to tourist arrivals to Malaysia, namely Australia, Brunei, China, Indonesia, India, Japan, the Philippines, South Korea, and the United Kingdom. Box-Jenkins time series method and Singular Spectrum Analysis are conducted and compared to study the best model to forecast the foreign tourist demand to Malaysia. Monthly data of tourism arrival in 1990 to 2014 were used and the forecasting were compared with 2015. Based on the results obtained, the forecasting model of Box-Jenkins time series method is the best model based on the percentage accuracy in forecasting the tourist demand to Malaysia.

Hierarchical pattern recognition for tourism demand forecasting

This study proposes a hierarchical pattern recognition method for tourism demand forecasting. The hierarchy consists of three tiers: the first tier recognizes the calendar pattern of tourism demand, identifying work days and holidays and integrating “floating holidays.” The second tier recognizes the tourism demand pattern in the data stream for different calendar pattern groups. The third tier generates forecasts of future tourism demand. Evidence from daily tourist visits to three attractions in China shows that the proposed method is effective in forecasting daily tourism demand. Moreover, the treatment of “floating holidays” turns out to be more effective and flexible than the commonly adopted dummy variable approach.

International tourism demand forecasting with machine learning models: The power of the number of lagged inputs

This study explores how to select the optimal number of lagged inputs (NLIs) in international tourism demand forecasting. With international tourist arrivals at 10 European countries, the performances of eight machine learning models are evaluated using different NLIs. The results show that: (1) as NLIs increases, the error of most machine learning models first decreases rapidly and then tends to be stable (or fluctuates around a certain value) when NLIs reaches a certain cutoff point. The cutoff point is related to 12 and its multiples. This trend is not affected by the size of the test set; (2) for nonlinear and ensemble models, it is better to select one cycle of the data as the NLIs, while for linear models, multiple cycles are a better choice; (3) significantly different prediction results are obtained by different categories of models when the optimal NLIs are used.

The role of disaggregated search data in improving tourism forecasts: Evidence from Sri Lanka

ABSTRACT Formulation of effective policies to enhance the resilience of tourism following the COVID-19 pandemic essentially requires comprehensive empirical information on changes in tourism demand and associated economic costs. The paper makes a novel contribution to tourism literature by employing regionally and temporally disaggregated tourism data and Google search data in improving the accuracy of tourism forecasts. Further, the paper adopts two timeseries variables namely tourist arrivals and guest nights in order to understand the changes due to COVID-19 in tourism demand more comprehensively. Monthly data on international tourist arrivals, guest nights and Google trends from 2004 to 2019 are used to produce regionally disaggregated (Europe, Asia, the Pacific, America, Other) monthly tourism forecasts for Sri Lanka. We find that SARMAX models outperform the other models (ARIMA, ARIMAX, SARIMA) in forecasting tourism demand following COVID-19. Interestingly, the paper makes a further step in utilizing forecasts in estimating foregone economic benefits due to COVID-19 pandemic. We find a notable difference in estimated direct economic loss depending on the variable used in estimates. The percentage loss is 40% when arrival forecasts are used in estimates and 29% when guest night forecasts are used in estimates. This provides important policy implications for improving post-COVID tourism.

Modeling and Forecasting of Tourist Arrivals in Crete Using Statistical Models and Models of Computational Intelligence

In the past few decades, tourism has clearly become one of the most prominent economic trends for many countries. For many destinations, this trend will continue to rise, and tourism will become the most dynamic and fastest growing sector of the economy. Thus, the reliable and accurate forecasting of tourism demand is necessary in making decisions for effective and efficient planning of tourism policy. The objective of this paper is the modeling and forecasting the international tourist arrivals to four prefectures of Crete in the year 2012, based on the actual tourist arrivals data over the period 1993 – 2011, using one-step-ahead forecast. In particular, this paper presented a comparative study of time series forecasts of international travel demand for the four prefectures of Crete using a variety of statistical quantitative forecasting models along with neural networks and fuzzy models.

우리나라 관광경제학 연구의 동향과 방향

This study aims to analyze trends in tourism economics research published to tourism and hospitality areas, and suggest research directions in the future. To this end this study searched for “tourism economics” from the database of KCI(Korea Citation Index) by Korean Research Foundation during the periods of 2002∼2020 and implemented language network analysis and visualization. Research papers of tourism economics were broadly classified by forecasting tourism demand, economic impacts, and valuation, in which major key words were derived. Results show that Journal of Tourism Sciences published the largest research papers of tourism economics(81), followed by Journal of Tourism & Leisure Research(77), Korean Journal of Hospitality & Tourism(46), Korean Journal of Tourism Research(39), International Journal of Tourism & Hospitality Research(39), and Journal of Hospitality & Tourism Studies(15). With regard to detailed fields of research papers, forecasting tourism demand was dominant, but showed a decreasing trend. Recently research papers of economic impacts indicated more increasing trend than valuation. The results of language network analysis indicate that major key words included “tourism”, “demand”, “input-output model”, “time-series”, “CVM”, “festival”, “willingness to pay(WTP)”, and “tourism demand.” The visualized results of language network show that the major key word of “tourism” was highlighted with “WTP”, “festival”, “tourist”, “multiplier”, and “time-series” during the periods of 2002∼2014. On the other hand, the major key word of “demand” was highlighted with “festival”, “tourism demand”, and “tourist“ during the periods of 2015∼2020. When comparing two periods, valuation papers were dominant in the past, but recently papers related to tourism demand and economic impacts showed an increasing trend. Furthermore, studies on festival were show a more increasing trend.

Forecasting Monthly Tourism Demand Using Enhanced Backpropagation Neural Network

The accurate forecasting of monthly tourism demand can improve tourism policies and planning. However, the complex nonlinear characteristics of monthly tourism demand complicate forecasting. This study proposes a novel approach named ICPSO-BPNN that combines improved chaotic particle swarm optimization (ICPSO) with backpropagation neural network (BPNN) to forecast monthly tourism demand. ICPSO with chaotic initialization and two search strategies, sigmoid-like inertia weight, and linear acceleration coefficients is utilized to search for the appropriate initial connection weights and thresholds necessary to improve the performance of BPNN. Two comparative real-life examples and one extended example are adopted to verify the superiority of the proposed ICPSO-BPNN. Results show ICPSO-BPNN outperforms that of the basic BPNN, autoregressive integrated moving average model, support vector regression, and other popular existing models.

“A study in grey”: a critical reflection on scientific (mis)behaviors in quantitative inquiries

ABSTRACT Application of quantitative techniques in the study of tourism phenomena has gained momentum in the last two decades. Various quantitative methods allow researchers to investigate problems encompassing, for instance, tourists’ behaviors, destinations efficiency, or tourism demand forecasting. However, statistical apparatus in the hands of a professional can either reveal a great deal about the world or conversely, deceive thousands of people. Past research touching upon the subject of scientific misbehaviors focused primarily on answering questions of who?, when? or how often?, leaving the matter of how specifically? rather uninvestigated. In the current paper, a grey area between quantitative data modification and falsification is discussed with the intention to build greater awareness among undergraduate students about potential pitfalls related to the execution of quantitative inquiries. To increase transparency and credibility of future scientific endeavors, practical recommendations are proposed on how to work responsibly within the confines of quantitative tourism research.

Tourism Forecasting and Tackling Fluctuating Patterns: A Composite Leading Indicator Approach

Abstract A growing interest is surfacing toward the non-linearities in tourism demand forecasting. This paper aims to construct a tourism composite indicator to anticipate the cyclical movement for the tourism demand in Fiji. The time duration tackled in the current study is under the span of approximately two decades from 2000 to 2017. Apart from an indicator construction approach and turning point dating algorithm, we proposed a two regime Markov switching model to analyse the Fijian tourism cycle. The empirical results revealed the composite leading indicator accorded a signalling approach for Fijian tourism demand with an average lead time of 2.75 months. Furthermore, the prior-recognition of the tourism economic transition with adequate dating evaluation of the tourism cycle is obtained from the filtered probabilities through the Markov switching models.

Machine Learning in Internet Search Query Selection for Tourism Forecasting

Prior studies have shown that Internet search query data have great potential to improve tourism forecasting. As such, selecting the most relevant information from large amounts of search query data is crucial to enhancing forecasting accuracy and reducing overfitting; however, such feature selection methods have not been considered in the tourism forecasting literature. This study employs four machine learning–based feature selection methods to extract useful search query data and construct relevant econometric models. We examined the proposed methods based on monthly forecasting of tourist arrivals in Beijing, China, along with weekly forecasting of hotel occupancy in the city of Charleston, South Carolina, USA. Our findings indicate that the forecasting model with the selected search keywords outperformed the benchmark ARMAX model without feature selection in forecasting tourism demand and hotel occupancy. Therefore, machine learning methods can identify the most useful search query data to significantly improve forecasting accuracy in tourism and hospitality.

Knowledge mapping of tourism demand forecasting research

Utilizing a scientometric review of global trends and structure from 388 bibliographic records over two decades (1999–2018), this study seeks to advance the building of comprehensive knowledge maps that draw upon global travel demand studies. The study, using the techniques of co-citation analysis, collaboration network and emerging trends analysis, identified major disciplines that provide knowledge and theories for tourism demand forecasting, many trending research topics, the most critical countries, institutions, publications, and articles, and the most influential researchers. The increasing interest and output for big data and machine learning techniques in the field were visualized via comprehensive knowledge maps. This research provides meaningful guidance for researchers, operators and decision makers who wish to improve the accuracy of tourism demand forecasting.

Are Combined Tourism Forecasts Better at Minimizing Forecasting Errors?

This study, which was contracted by the European Commission and is geared towards easy replicability by practitioners, compares the accuracy of individual and combined approaches to forecasting tourism demand for the total European Union. The evaluation of the forecasting accuracies was performed recursively (i.e., based on expanding estimation windows) for eight quarterly periods spanning two years in order to check the stability of the outcomes during a changing macroeconomic environment. The study sample includes Eurostat data from January 2005 until August 2017, and out of sample forecasts were calculated for the last two years for three and six months ahead. The analysis of the out-of-sample forecasts for arrivals and overnights showed that forecast combinations taking the historical forecasting performance of individual approaches such as Autoregressive Integrated Moving Average (ARIMA) models, REGARIMA models with different trend variables, and Error Trend Seasonal (ETS) models into account deliver the best results.

Tourism Demand Forecasting: A Decomposed Deep Learning Approach

Tourism planners rely on accurate demand forecasting. However, despite numerous advancements, crucial methodological issues remain unaddressed. This study aims to further improve the modeling accuracy and advance the artificial intelligence (AI)-based tourism demand forecasting methods. Deep learning models that predict tourism demand are often highly complex and encounter overfitting, which is mainly caused by two underlying problems: (1) access to limited data volumes and (2) additional explanatory variable requirement. To address these issues, we use a decomposition method that achieves high accuracy in short- and long-term AI-based forecasting models. The proposed method effectively decomposes the data and increases accuracy without additional data requirement. In conclusion, this study alleviates the overfitting issue and provides a methodological contribution by proposing a highly accurate deep learning method for AI-based tourism demand modeling.

Forecasting tourism demand using fractional grey prediction models with Fourier series

Tourism demand forecasting has played an important role in supporting governments to devise development policies for travel and tourism. However, time series related to tourism often do not conform to statistical assumptions and feature significant temporal fluctuations. Because a Fourier series is often applied to oscillating sequences to remove noise, it is reasonable to develop a grey prediction model in conjunction with a Fourier series to forecast tourism demand. However, grey prediction models traditionally use one-order accumulation, treating each sample with equal weight, to identify regularities concealed in data sequences. Furthermore, when generating residuals from Fourier series, the prediction accuracy of the newly generated predicted values is not taken into account. In this study, by using fractional order accumulation to assign appropriate weights to samples, we propose a fractional grey prediction model with Fourier series that offers high prediction accuracy. Experimental results demonstrate that the proposed grey prediction model performs well compared with other considered prediction models.