Specifying Smooth Transition Regression Models in the Presence of Conditional Heteroskedasticity of Unknown Form

Efthymios G Pavlidis 1 , 1 , Ivan Paya 2 , 2  and David A Peel 3 , 3
  • 1 Lancaster University Management School, e.pavlidis@lancaster.ac.uk
  • 2 Lancaster University Management School, i.paya@lancaster.ac.uk
  • 3 Lancaster University Management School, d.peel@lancaster.ac.uk

The specification of Smooth Transition Regression models consists of a sequence of tests, which are typically based on the assumption of i.i.d. errors. In this paper we examine the impact of conditional heteroskedasticity and investigate the performance of several heteroskedasticity robust versions. Simulation evidence indicates that conventional tests can frequently result in finding spurious nonlinearity. Conversely, when the true process is nonlinear in mean, the tests appear to have low size adjusted power and can lead to the selection of misspecified models. The above deficiencies also hold for tests based on Heteroskedasticity Consistent Covariance Matrix Estimators but not for the Fixed Design Wild Bootstrap. We highlight the importance of robust inference through empirical applications.

    • Data and code
Purchase article
Get instant unlimited access to the article.
Log in
Already have access? Please log in.

Log in with your institution

Journal + Issues

SNDE recognizes that advances in statistics and dynamical systems theory can increase our understanding of economic and financial markets. The journal seeks both theoretical and applied papers that characterize and motivate nonlinear phenomena. Researchers are required to assist replication of empirical results by providing copies of data and programs online. Algorithms and rapid communications are also published.