Qtools (version 1.5.8)

KhmaladzeFormat: Khmaladze Test

Description

This function provides significance levels of the Khmaladze Test using a (hard-coded) table of asymptotic critical values.

Usage

KhmaladzeFormat(object, epsilon)

Arguments

object

an object of class "KhmaladzeTest".

epsilon

trimming value. One of c(0.05,0.10,0.15,0.20,0.25,0.30).

Author

Marco Geraci

Details

This function is applied to an object produced by KhmaladzeTest. The Khmaladze test is used to test for location--shift and location-scale--shift hypotheses (Koenker, 2005). The test statistic is computed over the interval [epsilon, 1 - epsilon], where epsilon is the trimming value.

References

Appendix B in Koenker R. Quantile regression. New York, NY: Cambridge University Press; 2005.

Koenker R. and Xiao Z. Inference on the quantile regression process. Avalilable at http://www.econ.uiuc.edu/~roger/research/inference/khmal6ap.pdf.

Examples

data(barro, package = "quantreg")
eps <- 0.05
kt <- quantreg::KhmaladzeTest( y.net ~ lgdp2 + fse2 + gedy2 + Iy2 + gcony2, 
		data = barro, taus = seq(.05,.95,by = .01), trim = c(eps, 1 - eps))
class(kt)
KhmaladzeFormat(kt, epsilon = eps)