Are Turkey's tourism markets converging? Evidence from the two-step LM and three-step RALS-LM unit root tests

Abstract

In this study, we aim to analyse whether Turkey's 14 major tourist source markets are converging by using monthly data over the period January 1996 to December 2012. To this aim, we use the recently developed two-step Lagrange multiplier (LM) and three-step residual augmented least squares-Lagrange multiplier (RALS-LM) unit root tests that allow for two structural breaks in data. The results indicate that 10 out of 14 markets are converging, meaning that tourism policies and strategies directed at these markets are successful. Furthermore, the break points correspond to the important political, social, natural and economic events such as crisis, earthquake, disease and terrorist attack.

Keywords:

• tourism markets
• tourism policy
• convergence
• structural breaks
• unit root test

Acknowledgements

We are immensely grateful to Dr Ming Meng for providing us the RATS codes of unit root tests. An earlier version of this paper was presented at the 10th International Academic Conference, held in 3–6 June 2014 in Vienna, Austria

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. A nuisance parameter is any parameter which is not of immediate interest but which must be accounted for in the analysis of those parameters which are of interest.

2. A detailed explanation of this research strand could be provided in a new study.

3. Our definition of visitor arrivals involves only foreign visitors instead of citizens living abroad.

4. Critical values change based on the values, and therefore we have different critical values for each significance level for each country that has different values. The critical values were obtained for 200 observations via Monte Carlo simulations using 100,000 replications using the RATS code written by Dr Ming Meng. However, the critical values were also reported in Lee et al. (2012) and Meng and Lee (2012) for the LM and RALS-LM tests, respectively for different observation and break numbers.

5. R is taken as two in this study.