Testing for a Constant Coefficient of Variation in Nonparametric Regression
In this paper we propose a new test for the hypothesis of a constant coefficient of variation in the common nonparametric regression model. The test is based on an estimate of the L2-distance between the square of the regression function and the variance function. We prove asymptotic normality of a standardized estimate of this distance under the null hypothesis and fixed alternatives. The finite sample properties of a corresponding bootstrap test are investigated by means of a simulation study. The results are applicable to stationary processes with the common mixing conditions and are used to construct tests for ARCH assumptions in financial time series.
Key-wordsStationary processes Multiplicative error structure Generalized nonparametric regression models
AMS Subject Classification62G08 62G10 62G20
Unable to display preview. Download preview PDF.
- Carroll, R.J., Ruppert, D. 1988. Transformation and weighting in regression. Monographs on Statistics and Applied Probability, Chapman and Hall, New York.Google Scholar
- Dette, H., 2002. A consistent test for heteroscedasticity in nonparametric regression based on the kernel method. J. Stat. Plann.Inference, 103(1-2), 311-329.Google Scholar
- Dette, H., Hetzler, B., 2008. A simple test for the parametric form of the variance function in nonparametric regression. Ann. Inst. Stat. Math., To appear in:.Google Scholar
- Wieczorek, G., 2007. Tests auf multiplikative Struktur in strikt stationären Prozessen. Ph.D. thesis, Ruhr University Bochum (in German).Google Scholar