Affordable Access

The Use of Trimming to Improve the Performance of Tests for Nonlinear Serial Dependence with Application to the Australian National Electricity Market



In this article, we build on the results reported in Wild, Hinich and Foster (2008) for the National Electricity Market (NEM) of Australia by testing for episodic nonlinearity in the dynamics governing weekly cycles in spot price time series data. We apply the portmanteau correlation, bicorrelation and tricorrelation tests introduced in Hinich (1996) and the Engle (1982) ARCH LM test to the time series of half hourly spot prices from 7/12/1998 to 29/02/2008. We use trimming to improve the finite sample performance of the various test statistics mentioned above given the presence of significant skewness and leptokurtosis in the source datasets which may adversely affect the convergence properties of the test statistics in finite samples. With trimming, we still find the presence of significant third and fourth order (non-linear) serial dependence in the weekly spot price data, pointing to the presence of ‘deep’ nonlinear structure in this data.

There are no comments yet on this publication. Be the first to share your thoughts.