Keywords
international tourist arrivals, volatility, conditional correlation, seasonality, asymmetry, garch
Start Date
1-7-2004 12:00 AM
Abstract
International tourism is an important source of service exports to Spain and its regions, particularly the Canary Islands. Tourism is the major industry in the Canary Islands, accounting for about 22% of GDP. This paper examines time series of international tourism demand to the Canary Islands collected by the National Airport Administration (AENA) at airports from information regarding the number of tourist arrivals from abroad. The data set comprises monthly figures for the Canary Islands from 14 leading tourist source countries, as well as total tourist arrivals, from 1990(1)-2003(12). Tourist arrivals and associated volatilities for the monthly tourism data are estimated for the 14 source countries, as well as total tourist arrivals, using univariate and multivariate volatility models for the 15 data series. The univariate estimates suggest that the GARCH(1,1) model provides an accurate measure of conditional volatility in international monthly tourist arrivals for the 14 leading source countries, and total monthly tourist arrivals. The estimated conditional correlation coefficients provide useful information as to whether tourist source markets are similar in terms of shocks to international tourism demand. At the multivariate level, the conditional correlations in the shocks to monthly tourist arrivals are generally positive, varying from small negative to large positive correlations.
International Tourism Demand and Volatility Models for the Canary Islands
International tourism is an important source of service exports to Spain and its regions, particularly the Canary Islands. Tourism is the major industry in the Canary Islands, accounting for about 22% of GDP. This paper examines time series of international tourism demand to the Canary Islands collected by the National Airport Administration (AENA) at airports from information regarding the number of tourist arrivals from abroad. The data set comprises monthly figures for the Canary Islands from 14 leading tourist source countries, as well as total tourist arrivals, from 1990(1)-2003(12). Tourist arrivals and associated volatilities for the monthly tourism data are estimated for the 14 source countries, as well as total tourist arrivals, using univariate and multivariate volatility models for the 15 data series. The univariate estimates suggest that the GARCH(1,1) model provides an accurate measure of conditional volatility in international monthly tourist arrivals for the 14 leading source countries, and total monthly tourist arrivals. The estimated conditional correlation coefficients provide useful information as to whether tourist source markets are similar in terms of shocks to international tourism demand. At the multivariate level, the conditional correlations in the shocks to monthly tourist arrivals are generally positive, varying from small negative to large positive correlations.
