Tourism Demand Forecasting under Uncertainty by youmustknowme

VIEWS: 341 PAGES: 41

									Tourism Demand Forecasting under
          Uncertainty

                Prof. Haiyan Song


    School of Hotel and Tourism Management
     The Hong Kong Polytechnic University
                   Introduction
• Over the last three decades, numerous researchers have
  been involved in the area of tourism demand forecasting
  and a wide range of techniques, including quantitative
  (statistical) and qualitative (judgemental) approaches,
  have been developed.
• Statistical analyses can be carried out at different levels of
  sophistication, although most of them fall into two
  categories: econometric and time-series approaches (for
  detailed reviews of the tourism demand forecasting
  literature, see, for example, Witt and Witt (1995), Li et al.
  (2005), Song and Li (2008)).



                                                                   2
                       Introduction
• Statistical models can make efficient and optimal use of historic
  data and produce reliable forecasts if adequate data are available.
• In the meantime, statistical models also have the problem of the
  exclusive reliance on the availability and quality of historical time
  series. Current knowledge about special events (e.g., Olympics
  and Expos) or uncertainties that may happen in origins or
  destinations in the future is difficult to incorporate in such models.
• Judgemental methods, on the other hand, are often based on
  experts’ up-to-date knowledge about specific trends and events
  that can affect the variables being forecast.
• Despite their potential value, judgemental forecasts also have a
  number of serious disadvantages. For example, they are not
  consistent and are subject to subjective biases.




                                                                           3
                       Introduction
• Given the complementary strengths of statistical and judgemental
  methods, researchers have tried to integrate these two forecasting
  methods (Song and Li, 2008) with a view to improving forecasting
  accuracy.
• Research on the integration of statistical and judgemental forecasting
  methods was initiated as long as three decades ago and the benefits
  of such combination have been documented in many published
  studies.
• However, the design of forecasting systems that facilitate such
  integration has not attracted much attention (Goodwin, 2002).
• In tourism demand forecasting, studies on the integration of
  quantitative and qualitative techniques have been very limited.




                                                                           4
                 Objective
• The objective of this research is to look at
  how the integrative forecasting approach
  (combination of statistical and judgemental
  forecasts) can be used to forecast tourism
  demand
• Beijing and Shanghai will be used as
  examples.



                                                 5
                  Methodology
                – Method of integration
• A number of approaches have been proposed by previous
  researchers in terms of integrating judgement and
  statistical forecasts. Armstrong (1998) identifies the
  following five methods:
   – revising judgemental forecasts after seeing statistical
      extrapolations,
   – integrating mechanically the judgemental and statistical
      forecasts using appropriate formulas
   – using judgement to revise statistical extrapolations,
   – using a rule-based method to structurally adjust the
      statistical forecasting procedures, and
   – using judgemental inputs to identify a regression model.


                                                                6
                        Methodology
                      – Method of integration
•   According to Armstrong (1998), the third procedure listed above, i.e. using
    judgement to revise statistical extrapolations, is the most commonly used
    method.
•   This method is especially useful and likely to have a particular high payoff
    when forecasters are domain experts armed with up-to-date information
    and knowledge of future changes.
•   Specifically, the tourism demand forecasts are produced in two stages:
     – In the first stage, quantitative methods are applied to historical data to
        produce statistical forecasts;
     – In the second stage, the statistical extrapolations are revised based on the
        experts’ knowledge.




                                                                                      7
                     Methodology
       – Statistical method used in the first stage
• Modern econometric methods:
   – Standard economic theory suggests the most important
     factors that influence demand for tourism are:
        own price of tourism product,
        price of substitute tourism products,
        tourists’ income,
        tourism marketing expenditure,
        travel costs from origin countries/regions to the destination,
        one-off socio-economic events, etc.

Qi  f ( Pi , Pis , X i , M i , Ci , dummies  i )
                                           ,

                                                                          8
                  Methodology
         – Statistical method used in the first stage

    • Estimate the econometric models with a view to
      discover the economic relationship between
      tourism demand and its influencing factors
;     – The model used is known as dynamic econometric
        models
    • Calculate the demand elasticities to provide
      information for policy evaluation purpose




                                                         9
                    Methodology
      – Judgemental method used in the second stage
    • The second stage of the forecasting process
      consists of a Delphi-type revision through a panel
      of tourism experts.
    • Delphi method allows contributions from a group of
;
      experts who may be geographically dispersed.
      Most importantly, the Delphi approach is more
      accurate than traditional group meetings (Rowe
      and Wright, 1999).
    • In this study, the Delphi experts are selected from
      the academic institutions and government agencies


                                                            10
Beijing 2008 Olympics, China




                               11
            Background

The previous three Olympics

 Athens 2004, Greece
 Sydney 2000, Australia
 Atlanta 1996, USA




                              12
                           Athens 2004, Greece

                              Non-resident tourist arrivals
           18,000
Thousand




           16,000
           14,000
           12,000
           10,000
            8,000
            6,000
            4,000
            2,000
                0
                    1994

                           1995

                                  1996

                                         1997

                                                1998

                                                       1999

                                                              2000

                                                                     2001

                                                                            2002

                                                                                   2003

                                                                                          2004

                                                                                                 2005

                                                                                                        2006
                                                                                                               13
                          Sydney 2000, Australia

                           Non-resident visitor arrivals

           6,000
Thousand




           5,000

           4,000

           3,000

           2,000

           1,000

              0
                   1994

                          1995

                                 1996

                                        1997

                                               1998

                                                      1999

                                                             2000

                                                                    2001

                                                                           2002

                                                                                  2003

                                                                                         2004

                                                                                                2005

                                                                                                       2006
                                                                                                              14
                                  Atlanta 1996, USA

                              Non-resident tourist arrivals
           60,000
Thousand




           50,000

           40,000

           30,000

           20,000

           10,000

               0
                    1992

                           1993

                                  1994

                                         1995

                                                1996

                                                       1997

                                                              1998

                                                                     1999

                                                                            2000

                                                                                   2001

                                                                                          2002

                                                                                                 2003

                                                                                                        2004

                                                                                                               2005

                                                                                                                      2006
                                                                                                                             15
Beijing tourism demand analysis

               tourist arrivals          tourism revenues

             domestic   international   domestic   international


 income
              0.274        2.935         1.369        1.152
Elasticity

  price
              -0.878        n.a.          n.a.         n.a.
elasticity


                                                              16
Tourism demand forecasting
         - Beijing




                             17
              Beijing tourism demand forecasting
  T housand
 250,000

 200,000

 150,000

 100,000                                                                                                           Domestic tourist arrivals
  50,000

       0
              1994

                     1996

                            1998

                                   2000

                                          2002

                                                 2004

                                                        2006

                                                               2008

                                                                          2010

                                                                                    2012

                                                                                             2014

                                                                                                     2016
                               ARIMA forecasts                 Expert forecasts



                                                               T housand
                                                                7,000
                                                               6,000

                                                               5,000
                                                               4,000

                                                               3,000
International tourist arrivals                                 2,000
                                                               1,000
                                                                      0
                                                                             1978

                                                                                      1981

                                                                                              1984

                                                                                                     1987

                                                                                                            1990

                                                                                                                   1993

                                                                                                                          1996

                                                                                                                                 1999

                                                                                                                                        2002

                                                                                                                                               2005

                                                                                                                                                      2008

                                                                                                                                                              2011

                                                                                                                                                                          2014
                                                                                                        ARIMA forecasts                    Expert forecasts          18
                 Beijing tourism demand forecasting
Million RMB
350,000

300,000

250,000

200,000

150,000                                                                                                                Domestic tourist revenues
100,000

 50,000

      0
          1994

                  1996

                         1998

                                   2000

                                          2002

                                                   2004

                                                          2006

                                                                  2008

                                                                             2010

                                                                                     2012

                                                                                              2014

                                                                                                        2016
                                ARIMA fore casts                  Expe rt fore casts



                                                                 Million USD
                                                                 9,000
                                                                 8,000
                                                                 7,000
                                                                 6,000
                                                                 5,000
                                                                 4,000
International tourist revenues                                   3,000
                                                                 2,000
                                                                 1,000
                                                                         0
                                                                             1978

                                                                                    1981

                                                                                            1984

                                                                                                     1987

                                                                                                               1990

                                                                                                                      1993

                                                                                                                             1996

                                                                                                                                    1999

                                                                                                                                           2002

                                                                                                                                                  2005

                                                                                                                                                         2008

                                                                                                                                                                  2011

                                                                                                                                                                          2014
                                                                                                                                                                     19
                                                                                                       ARIMA fore casts                      Expe rt fore casts
         Part 2

Expo 2010 Shanghai, China




                            20
              Background
The first International Exhibition was held
 in London in 1851

Following the success of this event many
 highly successful exhibitions have been
 held over the last one and a half century
    For example, the Paris Exhibition of 1889 is
 well remembered for the creation of the Eiffel
 Tower

                                                   21
                  Background
•   International registered exhibition (world
    exhibition)
    •   Frequency : every five years
    •   Duration : 6 months at most
    •   Theme : general

 International recognised exhibition
    •   Frequency : during the interval between two
        international registered exhibitions
    •   Duration : 3 months at most
    •   Theme : specialized
 Horticultural international exhibitions
                                                      22
                        Background

                        Aichi       Hannove Kunming            Osaka
                         2005         2000         1999          1970
 Participating
                         121           155           95           75
  countries

 Total visitors       22 million    18 million   9.5 million   64 million

                     185 days       153 days     184 days     181 days
    Duration
                    (six months) (five months) (six months) (six months)
Average visitors         119           118         61.6           355
    per day           thousand      thousand     thousand      thousand
Data source: BIE official website
                                                                       23
                          Japan non-resident tourists arrivals
            8000
Thousands



            7000
            6000
            5000
            4000
            3000
            2000
            1000
                   1992

                           1993

                                  1994

                                         1995

                                                1996

                                                       1997

                                                                 1998

                                                                            1999

                                                                                   2000

                                                                                           2001

                                                                                                  2002

                                                                                                         2003

                                                                                                                2004

                                                                                                                       2005

                                                                                                                              2006
Data source: UNWTO yearbook
                                                                                          Gemany non-resident tourist arrivals

                                                                          25000
                                                              Thousands




                                                                          20000


                                                                          15000


                                                                          10000


                                                                           5000
                                                                                                                                             24
                                                                                    1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
                                  Kunming tourist arrivals
           800
Thousand




           700
           600
           500
           400
           300
           200
           100
             0
                 1986   1988   1990   1992   1994    1996      1998       2000      2002       2004      2006




                                                                                 Kunming domestic tourist arrivals ('000)

                                                2500
                                                2000
                                                1500
                                                1000
                                                    500
                                                      0
                                                       92

                                                              93

                                                                     94

                                                                            95

                                                                                   96

                                                                                          97

                                                                                                 98

                                                                                                        99

                                                                                                               00

                                                                                                                      01

                                                                                                                             02

                                                                                                                                    03

                                                                                                                                           04

                                                                                                                                                  05

                                                                                                                                                         06
                                                     19

                                                            19

                                                                   19

                                                                          19

                                                                                 19

                                                                                        19

                                                                                               19

                                                                                                      19

                                                                                                             20

                                                                                                                    20

                                                                                                                           20

                                                                                                                                  20

                                                                                                                                         20

                                                                                                                                                20

                                                                                                                                                       20
                                                                                                                                                       25
         Shanghai expo 2010

 Will be held from 01 May to 31 Oct 2010 (184
  days)
 First held in a developing country
 180 countries/regions has confirmed their
  participation by Oct 2008
 The official visitor forecast is 70 million (more
  than 64 million of Osaka Expo in 1970) among
  which, 95% are domestic visitors

                                                      26
Shanghai tourism demand analysis

               tourist arrivals          tourism revenues

             domestic   international   domestic   international


 income
              0.558        6.173         0.088         5.702
Elasticity

  price
              -1.152        n.a.         -0.190        n.a.
elasticity


                                                              27
      Shanghai tourism demand forecasting
    Thousand
    180,000

    160,000

    140,000

    120,000

    100,000

     80,000
                                                                                                                                                                                                Domestic tourist arrivals
     60,000

     40,000

     20,000

         0
              1997

                     1998

                            1999

                                   2000

                                          2001

                                                 2002

                                                        2003

                                                               2004

                                                                      2005

                                                                             2006

                                                                                    2007

                                                                                           2008

                                                                                                  2009

                                                                                                         2010

                                                                                                                2011

                                                                                                                       2012

                                                                                                                                     2013

                                                                                                                                              2014

                                                                                                                                                      2015

                                                                                                                                                             2016
                                                 ARIMA forecasts                                     Expert forecasts


                                                                                                                Thousand
                                                                                                                14,000


                                                                                                                12,000


                                                                                                                10,000


                                                                                                                 8,000


                                                                                                                 6,000
International tourist arrivals
                                                                                                                 4,000


                                                                                                                 2,000


                                                                                                                       0
                                                                                                                              1978

                                                                                                                                            1980

                                                                                                                                                     1982

                                                                                                                                                             1984

                                                                                                                                                                    1986

                                                                                                                                                                           1988

                                                                                                                                                                                  1990

                                                                                                                                                                                         1992

                                                                                                                                                                                                 1994

                                                                                                                                                                                                        1996

                                                                                                                                                                                                               1998

                                                                                                                                                                                                                      2000

                                                                                                                                                                                                                             2002

                                                                                                                                                                                                                                    2004

                                                                                                                                                                                                                                           2006

                                                                                                                                                                                                                                                  2008

                                                                                                                                                                                                                                                         2010

                                                                                                                                                                                                                                                                2012

                                                                                                                                                                                                                                                                       2014

                                                                                                                                                                                                                                                                              2016
                                                                                                                                                                                                                                                                       28
                                                                                                                                                                           ARIMA forecasts                                    Expert forecasts
          Shanghai tourism demand forecasting
 Million RMB
450,000
400,000
350,000
300,000
250,000
200,000                                                                                                    Domestic tourist revenues
150,000
100,000
 50,000
     0
          1999


                 2001


                        2003


                                 2005


                                           2007


                                                  2009


                                                              2011


                                                                            2013


                                                                                      2015
                        ARIMA fore casts                 Expe rt fore casts


                                                         Million USD
                                                         12,000

                                                         10,000

                                                          8,000

                                                          6,000

International tourist revenues                            4,000

                                                          2,000

                                                              0
                                                                     1980

                                                                              1983

                                                                                     1986

                                                                                             1989

                                                                                                    1992

                                                                                                            1995

                                                                                                                   1998

                                                                                                                          2001

                                                                                                                                 2004

                                                                                                                                        2007

                                                                                                                                                2010

                                                                                                                                                        2013

                                                                                                                                                               2016
                                                                                                                                                       29
                                                                                              ARIMA fore casts                   Expe rt fore casts
                           Findings
• The statistical and adjusted forecasts of total visitors show that the
  experts tend to be optimistic for both Beijing and Shanghai in terms of
  both visitor arrivals and revenues.

• The experts predict that the 2010 Olympics will benefit inbound
  tourism to Beijing so that the visitor growth in 2008 would be higher
  than the statistical forecasts.

• The growth of tourist arrivals will be higher than the revenue growth




                                                                            30
Integrating the forecasts through a
  Web-based forecasting system
                   System Design
                                      Forecasting Models       Tourism Demand
                                                              Forecasting System
                               Statistical                         (TDFS)
                               Forecasts
                    GUI
                  (Internet
    Academics     Browser)



                              HTML                                  Forecast
                              & Script                           Adjusting Tools

Public Sector                                 Web Server
                                                           Scenario Analysis Tools
                                             ADO &
                                             ODBC



                                              Tourism
  Practitioners                               Database



                                                                              32
33
34
35
36
   System Implementation
• Scenario analysis:




                           37
   System Implementation
• Changing the average growth rate of the tourism
  demand forecasts:




                                                    38
           Concluding Remarks
• The Web-based TDFS is developed and implemented as an
  innovative tool to integrate statistical methods and experts’
  adjustments in the tourism industry.
• The system enables the experts to make full use of statistical
  methods and adjust the statistical forecasts based on their
  specialized knowledge about the demand situations.
• Together with the advanced statistical forecasting techniques
  integrated in the TDFS, forecasting accuracy and reliability can be
  enhanced.
• Like other Web-based systems, the TDFS has four significant
  features – wide accessibility, flexibility, reusability, and user
  friendliness.




                                                                        39
        Concluding Remarks
• Delphi experts at various locations can visit the
  Website and make real-time judgmental
  adjustments to the statistical forecasts of
  tourism demand through Web browsers.

• Also, this system makes it easier to perform
  “what-if” scenario analyses on tourism demand
  forecasts, which can be very useful for
  policymakers and industry leaders for policy
  evaluation and decision making purposes.


                                                      40
Thank You!
hmsong@polyu.edu.hk

								
To top