Testing for Threshold Effects in the TARMA Framework

We present supremum Lagrange Multiplier tests to compare a linear ARMA specification against its threshold extension. derive the asymptotic distribution of test statistics both under null hypothesis and contiguous local alternatives. Moreover, we prove consistency tests. The Monte Carlo study shows that enjoy good finite-sample properties, are robust model mis-specification their performance is not affected if order unknown. low computational burden do suffer from some drawbacks affect quasi-likelihood ratio setting. Lastly, apply our time series standardized tree-ring growth indexes this can lead new research in climate studies.