ABSTRACT
The study examined the nexus between tourism and environmental pollution for six economies of South Asia, using a multivariate framework over the period 1995–2019. We employed Time-Varying Granger causality (TVGC) tests, non-parametric techniques, such as Marginal Adapted Regression Splines (MARS) and Generalized Additive Model (GAM). The TVGC results failed to provide evidence in support of causality between tourism and environmental pollution in all South Asian countries except Bhutan. However, the MARS method suggests significant evidence of tourism on environmental pollution in the non-parametric framework for India only. Furthermore, the GAM-based results support a significant relation between tourism and environmental pollution in Sri Lanka and Pakistan. Hence, it is highlighted that no method is best which can be applied for all countries and, therefore, researchers need to be careful in selecting an appropriate methodology to address the research questions. The findings of this study have provided some important suggestions to policy-makers.
Acknowledgements
We are thankful to two anonymous referees whose comments have improved the paper considerably. All remaining errors are solely our responsibility.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 United Nations World Tourism Organization.
2 World Travel and Tourism council.
3 Also see Smeral (Citation1996), Gössling et al. (Citation2005)
4 Four predictors are used to estimate MARS model because FDI is negative.