Carey Goh’s research while affiliated with Chinese University of Hong Kong and other places

Research on modeling the estimation and forecasting of tourism demand has evolved with increasing sophistication and improved quality. In this study, 155 research papers published between 1995 and 2009 were identified and were classified into three main groups according to the methods and techniques adopted—an econometric-based approach, time series techniques, and artificial intelligence (AI)-based methods. It appears that the more advanced methods such as cointegration, error correction model, time varying parameter model, and their combinations with systems of equations produce better results in terms of forecasting accuracy. We also discuss the implications and suggest future directions of tourism research techniques and methods.

The introduction of the Individual Visit Scheme (IVS), which allows Mainland Chinese travelers (hereafter called Chinese travelers) to visit Hong Kong on an individual basis, has been beneficial to the fast recovery of the local economy in Hong Kong from severe acute respiratory syndrome (SARS). The IVS travelers, together with the packaged travelers who visited Hong Kong by joining packaged tours, are presently forming the largest segment of inbound travelers to Hong Kong. Despite the significant economic contributions of Chinese travelers to Hong Kong in general, and to the foodservice sector in particular, there has been an absence of prior studies on IVS travelers’ behavior of selecting restaurants in the existing hospitality and tourism literature. This paper reports on an exploratory study that investigated the perceived importance of attributes that pertain to the selection of restaurants from the perspective of IVS and packaged travelers from Mainland China. Empirical findings of a questionnaire survey with 230 Chinese travelers, including 127 IVS travelers and 103 packaged travelers, showed that the respondents in general viewed the included attributes as relatively important. In addition, only two attributes exhibited significant differences between IVS and packaged travelers. Findings of this research would be useful for hospitality and tourism practitioners to better prepare for receiving the growing number of Chinese travelers.

This article reports a study that applies the rough sets algorithm to tourism demand analysis. Empirical outcomes are a set of automated but practical decision rules for practitioners from data that have a high degree of vagueness. We also introduce two new measures of qualitative noneconomic factors, namely a leisure time index and climate index into the forecasting framework. On the basis of long-haul U.S. and U.K. tourism demand for Hong Kong, empirical results show that leisure time and climate have stronger impacts on tourist arrivals than economic factors. Comprehensible decision rules are generated and tourism demand forecasts attain an accuracy of up to 80%. The findings put forward the importance of qualitative non- economic factors in travel motivation theory and demand analysis.

Purpose The purpose of this paper is to bring a new scientific theory into practice in the hotel industry. Design/methodology/approach In‐depth open‐ended interviews were used with eight hospitality practitioners, consumers, and academic researchers. Findings The article lays out current practices of security and privacy measurements in the hospitality industry and points out the related limitations. More importantly, it provides insights to hotel operators on possible future applications of a new scientific invention and how it could help alleviate the limitations found in the existing security and privacy measures. Practical implications The paper raises the awareness of a new scientific breakthrough that sheds new lights to security and privacy strategies in hotel operations. Hotels, which are able to react quickly to gain the first‐mover‐advantage in leading the industry on the application of such technology, will certainly gain tremendous publicity, and more importantly will be the first to create a new level of confidence in the market on security and privacy strategies implemented in their daily hotel operations. Originality/value This paper communicates a recent scientific development and its possible applications in the hotel industry. In this way, it bridges the gap between a scientific invention and its real‐world application. It is expected to provide insights for hoteliers who are interested in technology applications.

Purpose This study aims to examine online room rates in Hong Kong hotels. It focuses on comparing and contrasting the lowest hotel room rates that are available to customers. Design/methodology/approach Eight distribution channels and 45 hotels in Hong Kong were examined for online room rates in a 13‐month period from 2005 to 2006. The eight distribution channels represented different nature of operations including both indirect and direct distribution channels. Findings Empirical findings showed the web sites of local travel agents and local reservation agents offered the lowest online room rates, and that indirect distribution channels offered lower room rates than direct distribution channels. Research limitations/implications A major limitation of this study is the geographic limitation of hotel selection. Originality/value Findings of the study are expected to provide insights for hoteliers to refine their online room rate strategy.

Despite numerous studies have applied various forecasting models to tourism demand analysis, data mining techniques have been largely overlooked by academic researchers in tourism forecasting prior to 1999. Based on our review of published articles in tourism journals that applied data mining techniques to tourism demand forecasting, we find that the application of data mining techniques are still at their infancy. This paper concludes with practical implications and future research areas.

Most of the existing studies on tourism demand forecasting apply economic models that use mathematical functions, which require many statistical assumptions and limitations. This paper presents a new approach that applies the rough sets theory to form a forecasting model for tourism demand. The objective of this research is to create patterns which are able to distinguish between the classes of arrivals in terms of volume, based upon differences in the characteristics in each arrival. The information about the arrivals was organized in an Information Table where the number of arrivals corresponds to condition attributes, and the classification was defined by a decision attribute that indicated the forecast categorical value of future arrivals. Utilizing Japanese arrivals data in Hong Kong, empirical results showed the induced decision rules could accurately forecast (86.5%) of the test data.

This research integrates the rough sets theory into tourism demand analysis. Originated from the area of Artificial Intelligence, the rough sets theory was introduced to reveal important structures and to classify objects. A rough sets approach can capture useful information from a set of raw hybrid data and discover knowledge from the data in a form of decision rules. This makes the rough sets approach a useful classification and pattern recognition technique. Because of its ability to accommodate hybrid data and its algorithms without rigorous theoretical and statistical assumptions, the theory could complement the orthodox demand framework. This paper introduces a new rough sets approach for deriving rules from an Information Table of tourist arrivals. The induced rules were able to forecast change in demand with 87% accuracy.

This paper presents the use of time series SARIMA and MARIMA with interventions in forecasting tourism demand using ten arrival series for Hong Kong. Augmented Dickey–Fuller tests indicated that all the series were seasonal nonstationary. Significant interventions such as relaxation of the issuance of out-bound visitors visas, the Asian financial crisis, the handover, and the bird flu epidemic were all empirically identified with significant test results and expected signs. The forecasts obtained using models that capture stochastic nonstationary seasonality and interventions, SARIMA and MARIMA with intervention analysis, are compared with other eight time series models and were found to have the highest accuracy.