Tourism Forecasting Research Articles

Forecasting tomorrow’s tourist

Purpose This study aims to present a very recent literature review on tourism demand forecasting based on 50 relevant articles published between 2013 and June 2016. Design/methodology/approach For searching the literature, the 50 most relevant articles according to Google Scholar ranking were selected and collected. Then, each of the articles were scrutinized according to three main dimensions: the method or technique used for analyzing data; the location of the study; and the covered timeframe. Findings The most widely used modeling technique continues to be time series, confirming a trend identified prior to 2011. Nevertheless, artificial intelligence techniques, and most notably neural networks, are clearly becoming more used in recent years for tourism forecasting. This is a relevant subject for journals related to other social sciences, such as Economics, and also tourism data constitute an excellent source for developing novel modeling techniques. Originality/value The present literature review offers recent insights on tourism forecasting scientific literature, providing evidences on current trends and revealing interesting research gaps.

Mobility, Modernity, and Abroad

Mobility, Modernity, and Abroad

Thailand tourism forecasting based on a hybrid of discrete wavelet decomposition and NARX neural network

Purpose – Accurate forecast of tourist arrivals is crucial for Thailand since the tourism industry is a major economic factor of the country. However, a nonstationarity, normally consisted in nonlinear tourism time series can seriously ruin the forecasting computation. The purpose of this paper is to propose a hybrid forecasting method, namely discrete wavelet decomposition (DWD)-NARX, which combines DWD and the nonlinear autoregressive neural network with exogenous input (NARX) to cope with such nonstationarity, as a consequence, improve the effectiveness of the demand-side management activities. Design/methodology/approach – According to DWD-NARX, wavelet decomposition is executed for efficiently extracting the hidden significant, temporal features contained in the nonstationary time series. Then, each extracted feature set at a particular resolution level along with a relative price as an exogenous input factor are fed into NARX for further forecasting. Finally, the forecasting results are reconstructed. Forecasting performance measures rely on mean absolute percentage error, mean absolute error as well as mean square error. Model overfitting avoidance is also considered. Findings – The results indicate the superiority of the DWD-NARX over other efficient related neural forecasters in the cases of high forecasting performance rate as well as competently coping with model overfitting. Research limitations/implications – The scope of this study is confined to Thailand tourist arrivals forecast based on short-term projection. To resolve such limitations, future research should aim to apply the generalization capability of DWD-NARX on other domains of managerial time series forecast under long-term projection environment. However, the exogenous input factor is to be empirically revised on domain-by-domain basis. Originality/value – Few works have been implemented either to handle the nonstationarity, consisted in nonlinear, unpredictable time series, or to achieve great success on finding an appropriate and effective exogenous forecasting input. This study applies DWD to attain efficient feature extraction; then, utilizes the competent forecaster, NARX. This would comprehensively and specifically deal with the nonstationarity difficulties at once. In addition, this study finds the effectiveness of simply using a relative price, generated based on six top-ranked original tourist countries as an exogenous forecasting input.

A Review on Tourism Demand Modeling and Forecasting of Asian Countries

Tourism is very important economic source for the country, tourists bring money into the hosting countries, and so most countries try to attract more tourists. This research uses the methodology of literature review of some articles to monitor the factors affecting the tourism demand to Asian countries as well as finding the forecasting model that has been used to determine the number of tourists. The study concludes that the factors are of two types, short-term and long-term. The long-term factors are “income, tourism price, tourism price substitute, travel cost, travel value and exchange rate”. The short-term factors are economic crisis, SARS and the Gulf War. The most accurate forecasting model was found to be seasonal, non-seasonal ARIMA model.

Tourism demand modelling and forecasting using data mining techniques in multivariate time series: a case study in Turkey

In this study multiple linear regression, multilayer perceptron (MLP) regression, and support vector regression (SVR) are used to make multivariate tourism forecasting for Turkey. This paper is a comparative study of data mining techniques based on multivariate regression modelling with monthly data points to forecast tourism demand; it focuses on Turkey. Both MLP and SVR methods are widely employed in the variety forecasting problems. Most of the previous research on tourism forecasting used univariate time series or a limited number of variables with mostly yearly or quarterly, and rarely monthly frequencies. However, the application of data mining techniques for multivariate forecasting in the context of tourism demand has not been widely explored. This paper differs from earlier research in two ways: 1) it proposes multivariate regression modelling with monthly data points to forecast tourism demand; and 2) it focuses on Turkey by using a dataset with the most recently accumulated (between January 1996 and Dec 2013) 67 time series with respect to Turkey and its top 26 major tourism clients. Comparison of forecasting performances in terms of relative absolute error (RAE) and root relative squared error (RRSE) measurements shows that the SVR model, with RAE $=$ 12.34% and RRSE $=$ 14.02%, gives a better performance. The results obtained in this study provide information for researchers interested in applying data mining techniques to tourism demand forecasting and help policy makers, government bodies, investors, and managers for their regularization, planning, and investments by way of accurate tourism demand forecasting.

A meta-analysis of international tourism demand forecasting and implications for practice

Numerous studies on tourism forecasting have now been published over the past five decades. However, no consensus has been reached in terms of which types of forecasting models tend to be more accurate and in which circumstances. This study uses meta-analysis to examine the relationships between the accuracy of different forecasting models, and the data characteristics and study features. By reviewing 65 studies published during the period 1980–2011, the meta-regression analysis shows that the origins of tourists, destination, time period, modeling method, data frequency, number of variables and their measures and sample size all significantly influence the accuracy of forecasting models. This study is the first attempt to pair forecasting models with the data characteristics and the tourism forecasting context. The results provide suggestions for the choice of appropriate forecasting methods in different forecasting settings.

The Accuracy of Tourism Forecasting and Data Characteristics: A Meta-Analytical Approach

Sound policies and efficient managerial decisions in a wide range of touristic endeavors rely on accurate forecasts, but despite decades of extensive research and hundreds of published studies on the topic, neither a single model nor a category of forecasting models have proven to be superior. This meta-analysis of nearly 2,000 published tourism forecasts provides strong evidence that data characteristics are associated with forecast accuracy, and that this relationship varies across different categories of forecasting methods. The findings indicate that data characteristics might provide reliable indications as to which models are likely to be more accurate for certain tourism forecasting tasks, resulting in a more efficient and less costly process for identifying the best tourism forecasting model for the task.

Integrating Fuzzy Theory, Genetic Algorithm and Neural Network in Tourism Forecasting

Integrating Fuzzy Theory, Genetic Algorithm and Neural Network in Tourism Forecasting

The Impact of Competition Conditions on New Market Entrants in Istanbul Hotel Industry: An Analyse by Using Five Forces of Competitive Position Model of M.Porter

It is stated in the literature that about 25 million people attended in tourism movement in 1950 throughout the world, after the following 60 years this quantity has increased to 900 million people. In this parallelism, also the income of tourism industry has increased from 2,1 billion USD to 940 billion USD during the same 60 years in world wide. According to the tourism forecast of World Trade Organization (WTO) the number of tourists will increase to 1,8 billion people and the income will increase to 2 trillion USD in world wide (wto.org). Addition to this forecast, World Travel and Tourism Council (WTTC) indicated that Turkey will be one of the most growing tourism destinations in the world (wttc.org). When it is thought on the point of these assessments, to get profit as much as possible from this future situation and to have competitive advantage versus the other competitors; Turkish Tourism Industry needs to be developed with new invesments and the realization and the success of new invesments in an industry is related with the barriers of that industry for the new entrants. So in this study the availability of İstanbul Tourism Industry for new entrats was tried to be determined by making interwiews with the 36 managers of the hotel companies in İstanbul. The questions of interviews were prepared as depending on the Competitive Position Model of Porter.

Forecasting Overseas Visitors to the UK Using Continuous Time and Autoregressive Fractional Integrated Moving Average Models with Discrete Data

This paper applies Gaussian estimation methods to continuous time models for modelling overseas visitors into the UK. The use of continuous time modelling is widely used in economics and finance but not in tourism forecasting. Using monthly data for 1986–2010, various continuous time models are estimated and compared to autoregressive integrated moving average (ARIMA) and autoregressive fractionally integrated moving average (ARFIMA) models. Dynamic forecasts are obtained over different periods. The empirical results show that the ARIMA model performs very well, but that the constant elasticity of variance (CEV) continuous time model has the lowest root mean squared error (RMSE) over a short period.

Time Varying Parameter Error Correction Model Approach To Forecasting Tourist Arrivals In South Africa

In the field of tourism forecasting, the application of time-varying parameters has been successful in forecasting arrivals taking into account the changing behaviour of tourists. This article uses quarterly data to forecast intercontinental tourism demand for a long-haul, developing destination (South Africa) by applying time-varying parameters (TVP) to single equation estimates, and comparing these with vector autoregressive results. Ex-ante forecasts are done for tourist arrivals from various continents, and forecasting accuracy is evaluated by determining the Mean Absolute Percentage Error (MAPE), the Mean Absolute Deviation/Mean ratio (MAD/mean), the percentage Root Mean Square Error (RMSE) and Theil’s inequality coefficient. The results show that TVP do not always outperform other forecasting techniques.

Forecasting Daily Air Arrivals in Mallorca Island Using Nearest Neighbour Methods

This paper investigates the feasibility of using different generalizations of the nearest neighbour method in a tourism forecasting problem. The method is widely employed in different fields of research but, inexplicably, it is practically unknown in tourism forecasting. The analysis is centred not only in knowing the exact value of arrivals (point prediction), but also in anticipating the direction of the sign movement (sign prediction). Furthermore, this study also offers further evidence on a subject scarcely treated in tourism economics: the searching of predictable non-linear dynamics. The results encourage the use of this technique and reveal the existence of a non-linear seasonal effect in the analysed tourism time series.

A Non-Linear Tourism Demand Forecast Combination Model

It has been demonstrated in the tourism literature that a combination of individual tourism forecasting models can provide better performance than individual forecasting models. However, the linear combination uses only inputs that have a linear correlation to the actual outputs. This paper proposes a non-linear combination method using multilayer perceptron neural networks (MLPNN), which can map the non-linear relationship between inputs and outputs. UK inbound tourism quarterly arrivals data by purpose of visit are used for this case study. The empirical results show that the proposed non-linear MLPNN combination model is robust, powerful and can provide better performance at predicting arrivals than linear combination models.

Forecasting tourist arrivals in South Africa

Purpose: The aim of this paper is to model and forecast tourism to South Africa from the country's main intercontinental tourism markets. These include Great Britain, Germany, the Netherlands, the United States of America and France. Problem investigated: Tourism to South Africa has grown substantially since the first democratic elections in 1994. It is currently the third largest industry in the country and a vital source of foreign exchange earnings. Tourist arrivals continue to grow annually, and have shown some resilience to a number of emerging market crises, including the terrorist attacks in the USA. Business success, marketing decisions, government's investment policy as well as macroeconomic policy are influenced by the accuracy of tourism forecasts, since the tourism product comprises a number of services that cannot be accumulated. Accurate forecasts of tourism demand are paramount to ensure the availability of such services when demanded. In addition, the seasonal nature of tourism leads to a pattern of excess capacity followed by shortage in capacity. Method: Since univariate time series modelling has proved to be a very successful method for forecasting tourist arrivals, it is also the method employed in this paper. The naïve model is tested against a standard ARIMA model, as well as the Holt-Winters exponential smoothing and seasonal-non-seasonal ARIMA models. Forecasting accuracy is assessed using the mean absolute percentage error, root mean square error and Theill's U of the various models. Monthly tourist arrivals from 1994 to 2006 are used in the analysis, and arrivals are forecasted for 2007. Findings: The results show that seasonal ARIMA models deliver the most accurate predictions of arrivals over three time horizons, namely three months, six months and 12 months. Value: This paper is the first tourist arrivals forecast using South African data for the country as a whole, and therefore it forms an interesting case study as a long haul and growing tourist destination. Conclusion: The univariate forecasts provide fairly accurate forecasts of tourist arrivals in South Africa, especially over the short run. As such, it is understandable why it remains a popular approach to forecast tourist arrivals. However, this method does not make provision for assessing the influence of external events and therefore its policy application is limited.

The tourism forecasting competition

Abstract We evaluate the performances of various methods for forecasting tourism data. The data used include 366 monthly series, 427 quarterly series and 518 annual series, all supplied to us by either tourism bodies or academics who had used them in previous tourism forecasting studies. The forecasting methods implemented in the competition are univariate and multivariate time series approaches, and econometric models. This forecasting competition differs from previous competitions in several ways: (i) we concentrate on tourism data only; (ii) we include approaches with explanatory variables; (iii) we evaluate the forecast interval coverage as well as the point forecast accuracy; (iv) we observe the effect of temporal aggregation on the forecasting accuracy; and (v) we consider the mean absolute scaled error as an alternative forecasting accuracy measure. We find that pure time series approaches provide more accurate forecasts for tourism data than models with explanatory variables. For seasonal data we implement three fully automated pure time series algorithms that generate accurate point forecasts, and two of these also produce forecast coverage probabilities which are satisfactorily close to the nominal rates. For annual data we find that Naive forecasts are hard to beat.

Beyond point forecasting: Evaluation of alternative prediction intervals for tourist arrivals

This paper evaluates the performances of prediction intervals generated from alternative time series models, in the context of tourism forecasting. The forecasting methods considered include the autoregressive (AR) model, the AR model using the bias-corrected bootstrap, seasonal ARIMA models, innovations state space models for exponential smoothing, and Harvey’s structural time series models. We use thirteen monthly time series for the number of tourist arrivals to Hong Kong and Australia. The mean coverage rates and widths of the alternative prediction intervals are evaluated in an empirical setting. It is found that all models produce satisfactory prediction intervals, except for the autoregressive model. In particular, those based on the bias-corrected bootstrap perform best in general, providing tight intervals with accurate coverage rates, especially when the forecast horizon is long.

A Seasonal ARIMA Model of Tourism Forecasting: The Case of Taiwan

This paper aims to determine suitable SARIMA models to forecast the monthly outbound tourism departures of three major destinations from Taiwan to Hong Kong, Japan and the USA, respectively. The HEGY test is used to identify the deterministic seasonality in the data. The mean absolute per cent error (MAPE) is used to measure forecast accuracy. A low MAPE demonstrates the adequacy of the fitted SARIMA models. The results indicate that all series with first non-seasonal difference are needed to obtain the deterministic trend in outbound tourism series in Taiwan.

Bias‐corrected bootstrap prediction intervals for autoregressive model: new alternatives with applications to tourism forecasting

AbstractThis paper proposes the use of the bias‐corrected bootstrap for interval forecasting of an autoregressive time series with an arbitrary number of deterministic components. We use the bias‐corrected bootstrap based on two alternative bias‐correction methods: the bootstrap and an analytic formula based on asymptotic expansion. We also propose a new stationarity‐correction method, based on stable spectral factorization, as an alternative to Kilian's method exclusively used in past studies. A Monte Carlo experiment is conducted to compare small‐sample properties of prediction intervals. The results show that the bias‐corrected bootstrap prediction intervals proposed in this paper exhibit desirable small‐sample properties. It is also found that the bootstrap bias‐corrected prediction intervals based on stable spectral factorization are tighter and more stable than those based on Kilian's stationarity‐correction. The proposed methods are applied to interval forecasting for the number of tourist arrivals in Hong Kong. Copyright © 2010 John Wiley & Sons, Ltd.

Impacts of the Financial and Economic Crisis on Tourism in Asia

The purpose of this study is to forecast inbound tourism to Asia and outbound tourism from Asia by considering the possible impacts of the current financial and economic crisis. The autoregressive distributed lag model is used to calculate the interval demand elasticities, which are then used to generate interval forecasts for inbound tourism to and outbound tourism from Asia. Interval forecasts reduce the risk of complete forecasting failure arising from the uncertainties associated with the crisis. The forecast results suggest that the financial and economic crisis will have a negative impact on both inbound and outbound tourism in Asia but the demand will rebound from 2010.

Tourism forecasting using SARIMA models in Chilean regions

In this paper, we search for the best statistical model of time series with seasonal effect (seasonal autoregressive integrated moving average (SARIMA)) for forecasting arrivals in 13 regions of Chile. We use monthly time series of arrivals from January 2004 to March 2009. The forecasting performance is assessed using data for the period October 2008 to March 2009. We use three methods for the specification of the model: the Box?Jenkins method with Akaike criterion, the method of minimising the forecast error and the regARIMA method of the X12-ARIMA package. We compare the performance of the three methods according to their forecast results. Regions have different SARIMA specifications, resembling the underlying differences in tourism infrastructure and capacities available within each region.