The coefficient of determination, R2, which measures the variance in the data explained by the model, is used to assess the model’s ability to reproduce the dynamics within the tourism sector between 1990 and 2005. The parameters of interest are the number of tourist arrivals, the number of beds in the hotel sector, and the number of workers in the accommodation industry. Figures 1 - 3 show the actual versus the simulated data for the aforementioned parameters. Both the r (correlation coefficient) and the R2 are reasonably high for each of the variables, showing that the model captures the mean and the trends in the actual data quite well. It should be noted that the tourism arrivals sub-model could not/ did not 'predict' the down-turn in stayover visitors as a result of 9-11 but that an exogenous shock was added to replicate the impact the downturn had on arrivals to the island.10 Indeed, the knock-on effect of 9-11 on the labour market, bed capacity and occupancy rates served as a benchmark to validate the internal consistency of the model.
Annual Tourist Arrivals
r = 98.0
R2 = 96.9
SIMULATED ANNUAL ARRIVALS ANNUAL TOURIST ARRIVALS
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
10 The growth in tourist expenditure was impeded by the events of 11 September 2001 in the islands main tourist market, the United States of America. In an attempt to attract visitors to the Island, given the fall in world travel, hoteliers were compelled to offer discounted rates, which further reduced tourist revenues in the following year.