Integrating statistical and judgmental tourism demand forecasting approaches : the case of Hong Kong

Pao Yue-kong Library Electronic Theses Database

Integrating statistical and judgmental tourism demand forecasting approaches : the case of Hong Kong


Author: Lin, Shanshan
Title: Integrating statistical and judgmental tourism demand forecasting approaches : the case of Hong Kong
Degree: Ph.D.
Year: 2013
Subject: Tourism -- Forecasting.
Tourism -- China -- Hong Kong -- Forecasting.
Hong Kong Polytechnic University -- Dissertations
Department: School of Hotel and Tourism Management
Pages: xv, 442 p. : ill. ; 30 cm.
Language: English
InnoPac Record:
Abstract: In the past 4 decades, quantitative forecasting methods have overwhelmingly dominated tourism demand forecasting studies, while qualitative forecasting research has been correspondingly rare, even though judgmental forecasting has often been exercised in many tourism businesses as a routine and on an informal basis. Given the respective strengths and weaknesses of quantitative and judgmental forecasting methods, it seems sensible to integrate them by combining information from multiple sources. Although combining forecasts has attracted broad attention in the general forecasting literature, only a few studies on this topic have appeared in the tourism forecasting literature. Perhaps due to the absence of contextual information in the forecasting and combination process, the integration of statistical forecasts has led to limited improvement in accuracy. Thus, future studies should focus on integrating judgmental input (including contextual information) into statistical forecasts. To date, there has been little research that has comprehensively examined the effectiveness of integrating judgmental and statistical forecasting methods in the tourism context. Compared to the extensive research on the integration of forecasting techniques in other fields, there is a significant gap in the tourism literature. This study presents the first attempt to develop a research framework for the integration of econometric and judgmental forecasts based on Hong Kong tourism demand data with a view to providing recommendations and suggestions for decision-makers in both the public and private sectors in Hong Kong. The quarterly forecasts of visitor arrivals in Hong Kong from 6 source markets (i.e. Mainland China, Taiwan, Japan, the USA, the UK, and Australia) up to 2015 were generated using an econometric model, namely the autoregressive distributed lag model -error correction model (ARDL-ECM). The incorporation of experts' domain knowledge into the statistical forecasts by utilizing the Delphi technique via a forecasting support system (Hong Kong Tourism Demand Forecasting System, HKTDFS) improved forecast accuracy of the integration framework. A qualified panel of experts was formed; the panel members were different stakeholders from the government, accommodation, and tourist attraction sectors, and academics from various institutions.
To establish a holistic analytical framework for integrating statistical forecasts with human judgment, both quantitative and qualitative analyses were applied. The quantitative analysis aimed to examine the forecasting performance of statistical and judgmental forecasts from 3 dimensions: accuracy, bias, and efficiency. The research hypotheses were tested by examining the values of the error measures, conducting correlation and regression analyses, and employing statistical tests. Comparisons were made to examine the difference in accuracy among different Delphi rounds, source markets, expert groups, expertise levels, levels of data variability, forecasting horizons, and sizes and directions of adjustments. The findings suggested that, on average, statistical forecasts adjusted by the Delphi experts improved forecast accuracy for all of the 6 markets. The results showed that the consensus group forecasts in the final round of the Delphi survey provided significantly more accurate forecasts than those of the initial statistical forecasts and the simple average of individual experts’ forecasts in Round 1. Although satisfactory accuracy was achieved, the group forecasts were found to be inefficient and biased for some of the individual markets. It was also found that the industry experts performed better than the academic experts, indicating the value of incorporating contextual knowledge into statistical forecasts. In-depth interviews were conducted to provide qualitative input to interpret the quantitative findings from the hypothesis tests, examine the underlying assumptions embodied in the experts' forecasting adjustment process, and collect experts' opinions regarding the use of the forecasting system to aid their judgmental adjustments. The interview findings confirmed that compared to the academic experts, the industry experts preferred to use simpler and easier forecasting methods. The experts reached the consensus that given the relative strengths and weaknesses of judgmental and statistical forecasting methods, it is necessary to integrate these 2 types of forecasts in order to make better tourism demand forecasts. According to the experts interviewed, a variety of reasons were identified as being responsible for the accuracy improvement in this study, such as the provision of multiple information cues (e.g. time-series information and non-time series cues), the use of a Web-based forecasting support system, and the use of the Delphi technique to structure and aggregate experts' judgments. Useful recommendations and suggestions were made by the experts to further improve the HKTDFS and to point to future research directions.

Files in this item

Files Size Format
b26527741.pdf 4.696Mb PDF
Copyright Undertaking
As a bona fide Library user, I declare that:
  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.


Quick Search


More Information