Composite quantile regression estimation in non-parametric regression model under left-truncated data
WANG Jiang-feng1, TIAN Xiao-min2, ZHANG Hui-zeng2, WEN Li-min3
1. School of Statis. Math., Zhejiang Gongshang Univ., Hangzhou 310018, China
2. Dept. of Math., Hangzhou Normal Univ., Hangzhou 310036, China
3. School of Math. Sci., Jiangxi Normal Univ., Nanchang 330022, China
School of Inform. Tech., Jiangxi Normal Univ., Nanchang 330013, China
Abstract In this paper, a local linear composite quantile regression estimator of regression function is constructed in the regression model with heteroscedastic error under left-truncated data. The asymptotic normality of the proposed estimator is also established. The estimator is much more efficient than the local linear regression estimator for commonly-used non-normal error distributions via simulations.
WANG Jiang-feng, TIAN Xiao-min, ZHANG Hui-zeng, WEN Li-min. Composite quantile regression estimation in non-parametric regression model under left-truncated data. Applied Mathematics A Journal of Chinese Universities, 2015, 30(1): 71-83.