Abstract

This paper examines the power properties of several linearity tests applied in time-series analysis. The tests are the ones Lee et al.(1993) used in their Monte Carlo study. The main tool used for power comparisons in this paper is the Pitman asymptotic relative efficiency. The results generally strengthen the outcome of the simulations and complement some results in Lee et al. (1993). They also suggest guidelines for designing Monte Carlo experiments for linearity tests.

Recommended Citation

Timo Teräsvirta (1996) "Power Properties of Linearity Tests for Time Series", Studies in Nonlinear Dynamics & Econometrics: Vol. 1: No. 1, Article 2.
http://www.bepress.com/snde/vol1/iss1/art2