Demand volatility analysis in turkish tourism industry using multivariate GARCH (MGARCH) models

Tourism is a significant industry for developing countries like Turkey, where it supports economic growth in terms of balance of payments, national income and employment. And tourism demand is a priority issue for the industry. Tourism demand is extremely sensitive to epidemic diseases, natural disasters and terrorism along with economical, financial and political problems. Therefore, investigating the changes in demand, or volatility is a major issue in order to determine the appropriate policies and strategies for tourism industry. The aim of this study is to determine the volatility in Turkey?s incoming tourism demand. First differences of the logarithms of tourist arrivals from the top four tourist-generating countries, namely Germany, France, the Netherlands and the United Kingdom, were analyzed by MGARCH models for the period 1985:01-2010:10. Three MGARCH parameterizations, diagonal VECH, diagonal BEKK, and CCC, were applied. Further analysis is conducted by CCC-MGARCH (1,1) model, as it produces the highest log-likelihood value and the lowest AIC, SIC and HQ values. Model results suggest that there are interactions between the selected countries. This result is important as it emphasizes the mutual dependency between the most significant tourist markets. All countries included in the analysis are affected by their past volatilities to some extent. Increase in the demand volatility of a country causes volatility increase for the other countries. This interdependency is strengthened by the fact that selected countries are affected more by the other countries than their own past shocks.