A Consistent Nonparametric Test for Causality in Quantile

(1999), we can derive the asymptotic variance which is based on the i.i.d. sequence having the
same marginal distributions as weakly dependent variables in the test statistic. With this little
trick we only need to show that the asymptotic variance is o(1) in an i.i.d. situation. For
details refer to the Appendix.

4. Conclusion

This paper has provided a consistent test for Granger-causality in quantile. The test can be
extended to testing conditional quantile restrictions with dependent data; for example, testing
misspecification test, testing the insignificance of a subset of regressors, testing some
semiparametric versus nonparametric models, all in quantile regression models.

<<   7   8   9   10   11   12   13   >>

More intriguing information