University of Helsinki
A working paper version: HECER Discussion Paper No 124 / October 2006.
PDF Document:
http://ethesis.helsinki.fi/julkaisut/eri/hecer/disc/124/misspeci.pdf
Abstract: We develop misspeciĀ cation tests based on quantile residuals and apply them to nonlinear time series models for which conventional residuals are not suited. We formulate a general framework and use it to obtain computationally simple tests aimed at detecting autocorrelation, conditional heteroscedasticity, and non-normality in quantile residuals. These tests can be viewed as generalizations of similar previous tests based on conventional residuals and the Lagrange Multiplier principle. According to simulations on mixture models the proposed tests have reasonable size properties and power in cases when the traditional approach does not work. An empirical example on macroeconometric data illustrates the usefulness of these methods.
Download GAUSS codes:
How to install Code for computing the tests Example code for estimating a mixture distribution based model Dataset for the example code