Results
Figure 2 shows the output of ForecastPro for
expenditures in restaurants
and hotels (RH).
Figure 2. Restaurant and Hotel expenditures, historical and forecasted.
Another estimator for tourism is the number of non-residents at Geor-
gia’s borders. The history of this series and a 5-year forecast is shown below.
The growth rate exceeds that of RH expenditures.
Figure 3. Non-resident arrivals at Georgia’s borders, historical and
forecasted.
34
Caucasus Journal of Social Sciences
Con idence Limits
Forecasts
Con idence Limits
History
Con idence Limits
Forecasts
Con idence Limits
History
Alternative scenario forecasts for tourism in Georgia are shown here for
2013.
Conclusion
The tourism study, as
well as the current literature, strongly suggest that
judgmental methods may improve the accuracy of forecasts when combined
in an orderly way with statistical forecasts. Scenario scripting is an effective
way to forecast under conditions of a high degree of uncertainty.
Recommendations for Further Research
Continuing research is needed in the application of forecasting method-
ologies in the countries of the former Soviet Union.
In light of skepticism regarding the reliability of data provided by gov-
ernments
of nations in transition, more research is needed in verifying the
reliability of sources.
Further research is needed in forecasting under the conditions of great
uncertainty that prevail in economies affected by unstable relations with near
neighbors and internal opposition.
The “Tourist Trauma Trough” should be studied in greater detail in order
to understand the depth and duration of periods of lost income in a variety of
situations.
More research is needed in combining both methods of forecasting and
results from different sources using the same method.
Combining statistical
and judgmental forecasts, in particular, needs more attention.
Forecasting accuracy in economies and sectors in post-Soviet states
needs to be monitored and analyzed over time.
Further research in tourism demand forecasting is needed in nations in
transition from authoritarian systems to open societies. Narayan (2003),
however, concludes his extensive literature review with the comment that
“…the tourism demand literature must be viewed with caution…” (p. 377).
There is an almost complete absence of research in the area of return-on-
investment in tourism in the post-Soviet space. This
is a most fertile area for
further study.
35
Caucasus Journal of Social Sciences
Best
Middle
Worst
Case
Case
Case
(B)
(M)
(W)
Expected Value Factor
1.20
1.00
0.07
Statistical Forecast RH (mil GEL)
524
Adjusted Forecast RH (mil GEL)
891
524
37
Statistical Forecast B (000 persons)
2,226
Adjusted Forecast B (000 persons)
2,671
2,226
156
References
Armstrong, J. S. (2006). Findings from evidence-based forecasting: Meth-
ods for reducing forecast error.
International Journal of Forecasting, 22(3),
583-598.
Armstrong, J. S. (1999). Introduction to paper and commentaries on the
Delphi technique.
International Journal of Forecasting, 15, 351-352.
Armstrong, J. S. (1989). Combining forecasts: The end of the beginning
or the beginning of the end?
International Journal of Forecasting, 5, 585-588.
Makridakis, S., & Hibon, M. (1979). Accuracy of Forecasting:
An Empirical
Investigation
Journal of the Royal Statistical Society. Series A (General), 142(2),
97-145
Raupp, E. R. (2009). Why do we need forecasts?
The Georgia Forecast.
Retrieved July 13, 2009, from http://tgf.ge.
Raupp, E. R., & Apkhazava, N. (2009). Lifting the veil: Forecasting tourism
in post-Soviet Georgia.
International Symposium on Forecasting, June 21-24,
Hong Kong.
Stellwagen, E. A., & Goodrich, R. L. (2008).
ForecastPro user’s guide. Bel-
mont, MA: Business Forecast Systems, Inc.
Taleb, N. N. (2007).
The black swan: The impact of the highly improbable.
New York: Random House.
36
Caucasus Journal of Social Sciences