Difference-based methods have attracted increasing attention for analyzing partially linear models in the recent literature. In this paper, we first propose to solve the optimal sequence selection problem in difference-based estimation for the linear component. To achieve the goal, a family of new sequences and a cross-validation method for selecting the adaptive sequence are proposed. We demonstrate that the existing sequences are only extreme cases in the proposed family. Secondly, we propose a new estimator for the residual variance by fitting a linear regression method to some difference-based estimators. Our proposed estimator achieves the asymptotic optimal rate of mean squared error. Simulation studies also demonstrate that our proposed estimator performs better than the existing estimator, especially when the sample size is small and the nonparametric function is rough.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
Tax calculation will be finalised during checkout.
Akdeniz F, Duran E (2013) New difference-based estimator of parameters in semiparametric regression models. J Stat Comput Simul 83:810–824
Aneiros G, Ling N, Vieu P (2015) Error variance estimation in semi-functional partially linear regression models. J Nonparametr Stat 27:316–330
Chen H, Shiau J (1991) A two-stage spline smoothing method for partially linear models. J Stat Plan Inference 27:187–201
Cuzick J (1992) Semiparametric additive regression. J Roy Stat Soc B 54:831–843
Dai W, Ma Y, Tong T, Zhu L (2015) Difference-based variance estimation in nonparametric regression with repeated measurement data. J Stat Plan Inference 163:1–20
Dai W, Tong T, Genton M (2016) Optimal estimation of derivatives in nonparametric regression. J Mach Learn Res 17:1–25
Dai W, Tong T, Zhu L (2017) On the choice of difference sequence in a unified framework for variance estimation in nonparametric regression. Stat Sci 32:455–468
Dette H, Munk A, Wagner T (1998) Estimating the variance in nonparametric regression—what is a reasonable choice? J Roy Stat Soc B 60:751–764
Eubank R, Kambour E, Kim J, Klipple K, Reese C, Schimek M (1998) Estimation in partially linear models. Comput Stat Data Anal 29:27–34
Fan J, Huang T (2005) Profile likelihood inferences on semiparametric varying-coefficient partially linear models. Bernoulli 11:1031–1057
Gasser T, Sroka L, Jennen-Steinmetz C (1986) Residual variance and residual pattern in nonlinear regression. Biometrika 73:625–633
Hall P, Kay JW, Titterington DM (1990) Asymptotically optimal difference-based estimation of variance in nonparametric regression. Biometrika 77:521–528
Hall P, Keilegom IV (2003) Using difference-based methods for inference in nonparametric regression with time series errors. J Roy Stat Soc B 65:443–456
Hardle W, Liang H, Gao J (2000) Partially linear models. Physika Verlag, Heidelberg
He H, Tang W, Zuo G (2014) Statistical inference in the partial inear models with the doubles moothing local linear regression method. J Stat Plan Inference 146:102–112
Hu H, Zhang Y, Pan X (2016) Asymptotic normality of dhd estimators in a partially linear model. Stat Pap 57:567–587
Levine M (2015) Minimax rate of convergence for an estimator of the functional component in a semiparametric multivariate partially linear model. J Multivar Anal 140:283–290
Liu Q, Zhao G (2012) A comparison of estimation methods for partially linear models. Int J Innov Manag Inf Prod 3:38–42
Lokshin M (2006) Difference-based semiparametric estimation of partial linear regression models. Stata J 6:377–383
Rice JA (1984) Bandwidth choice for nonparametric regression. Ann Stat 12:1215–1230
Schott JR (1997) Matrix analysis for statistics. Wiley, New York
Severini T, Wong W (1992) Generalized profile likelihhood and conditional parametric models. Ann Stat 20:1768–1802
Spechman P (1988) Kernel smoothing in partial linear models. J Roy Stat Soc B 50:413–436
Tabakan G (2013) Performance of the difference-based estimators in partially linear models. Statistics 47:329–347
Tong T, Ma Y, Wang Y (2013) Optimal variance estimation without estimating the meam function. Bernoulli 19:1839–1854
Tong T, Wang Y (2005) Estimating residual variance in nonparametric regression using least squares. Biometrika 92:821–830
Wang L, Brown L, Cai T (2011) A difference based approach to the semiparametric partial linear model. Electron J Stat 5:619–641
Wang W, Lin L (2015) Derivative estimation based on difference sequence via locally weighted least squares regession. J Mach Learn Res 16:2617–2641
Wang W, Lin L, Yu L (2017) Optimal variance estimation based on lagged second-order difference in nonparametric regression. Comput Stat 32:1047–1063
Whittle P (1964) On the convergence to normality of quadratic forms in independent variables. Teor. Verojatnost. i Primenen 9:113–118
Yatchew A (1997) An elementary estimator of the partial linear model. Econ Lett 57:135–143
Zhao H, You J (2011) Difference based estimation for partially linear regression models with measurement errors. J Multivar Anal 102:1321–1338
Zhou Y, Cheng Y, Wang L, Tong T (2015) Optimal difference-based variance estimation in heteroscedastic nonparametric regression. Stat Sin 25:1377–1397
Yuejin Zhou’s research was supported in part by the Natural Science Foundation of Anhui Grant (No. KJ2017A087), and the National Natural Science Foundation of China Grant (No. 61472003). Yebin Cheng’s research was supported in part by the National Natural Science Foundation of China Grant (No. 11271241). Tiejun Tong’s research was supported in part by the Hong Kong Baptist University Grants FRG1/16-17/018 and FRG2/16-17/074, and the National Natural Science Foundation of China Grant (No. 11671338).
About this article
Cite this article
Zhou, Y., Cheng, Y., Dai, W. et al. Optimal difference-based estimation for partially linear models. Comput Stat 33, 863–885 (2018). https://doi.org/10.1007/s00180-017-0786-3
- Asymptotic normality
- Difference-based method
- Difference sequence
- Least squares estimator
- Partially linear model