Abstract
In this paper, two new tests for heteroscedasticity in nonparametric regression are presented and compared. The first of these tests consists in first estimating nonparametrically the unknown conditional variance function and then using a classical least-squares test for a general linear model to test whether this function is a constant. The second test is based on using an overall distance between a nonparametric estimator of the conditional variance function and a parametric estimator of the variance of the model under the assumption of homoscedasticity. A bootstrap algorithm is used to approximate the distribution of this test statistic. Extended versions of both procedures in two directions, first, in the context of dependent data, and second, in the case of testing if the variance function is a polynomial of a certain degree, are also described. A broad simulation study is carried out to illustrate the finite sample performance of both tests when the observations are independent and when they are dependent.
Similar content being viewed by others
References
Alcalá JT, Cristobal JA, González-Manteiga W (1999) Goodness-of-fit test for linear models based on local polynomials. Stat Probab Lett 42: 39–46
Altman N (1993) Estimating error correlation in nonparametric regression. Stat Probab Lett 18: 213–218
Davidson R, MacKinnon JG (1986) Graphical methods for investigating the size and power of hypothesis tests. Manchester Sch 66: 1–26
Dette H (2002) A consistent test for heteroscedasticity in nonparametric regression based on the kernel method. J Stat Plan Inference 103: 311–329
Dette H, Munk A (1998) Testing heteroscedasticity in nonparametric regression. J R Stat Soc Ser B 60: 693–708
Dette H, Neumeyer N, Van Keilegom I (2007) A new test for the parametric form of the variance function in nonparametric regression. J R Stat Soc Ser B 69: 903–917
Fan J, Gijbels I (1996) Local polynomial modeling and its applications. Chapman & Hall, London
Fan J, Yao Q (1998) Efficient estimation of conditional variance functions in stochastic regression. Biometrika 85: 645–660
Francisco-Fernández M, Opsomer JD, Vilar-Fern ández JM (2004) A plug-in bandwidth selector for local polynomial regression estimator with correlated errors. J Nonparametr Stat 16(1-2): 127–151
Francisco-Fernández M, Vilar-Fernández JM (2001) Local polynomial regression estimation with correlated errors. Commun Stat Theory Methods 30(7): 1271–1293
Gasser T, Sroka L, Jennen-Steinmetz C (1986) Residual variance and residual pattern in nonlinear regression. Biometrika 73: 625–633
Gijbels I, Rousson V (2001) A nonparametric least-squares test for checking a polynomial relationship. Stat Probab Lett 51: 253–261
Hall P, Kay JW, Titterington DM (1990) Asymptotically optimal difference-based estimation of variance in nonparametric regression. Biometrika 77: 521–528
Hall P, Van Keilegom I (2003) Using difference-based methods for inference in nonparametric regression with time-series errors. J R Stat Soc Ser B 65: 443–456
Härdle W, Mammen E (1993) Comparing nonparametric versus parametric regression fits. Ann Stat 21: 1926–1947
Liero H (2003) Testing homoscedasticity in nonparametric regression. J Nonparametr Stat 15: 31–51
Rice J (1984) Bandwidth choice for nonparametric regression. Ann Stat 12: 1215–1230
Vilar-Fernández JM, Francisco-Fernández M (2006) Nonparametric estimation of the conditional variance function with correlated errors. J Nonparametr Stat 18(4–6): 375–391
Vilar-Fernández JM, González-Manteiga W (2000) Resampling for checking linear regression models via non-parametric regression estimation. Comput Stat Data Anal 35: 211–231
Wang L, Zhou XH (2005) A fully nonparametric diagnostic test for homogeneity of variances. Can J Stat 33(4): 545–558
Zheng JX (1996) A consistent test of a functional form via nonparametric estimation techniques. J Econom 75: 263–289
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Francisco-Fernández, M., Vilar-Fernández, J.M. Two tests for heteroscedasticity in nonparametric regression. Comput Stat 24, 145–163 (2009). https://doi.org/10.1007/s00180-008-0110-3
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00180-008-0110-3