An alternative bandwidth selection method for estimating functional coefficient models
作者:Chen, XR(Chen, Xirong);Huang, TC(Huang, Ta-Cheng);Li, Q(Li, Qi)
ECONOMICS LETTERS
DOI:10.1016/j.econlet.2017.03.009
出版年:JUL 2017
摘要
Functionalcoefficientregressionmodelsare very usefulformany statistics and economics applications and there exists a large body of literature on kernel estimation of thecoefficientfunctions. Fan and Zhang (1999) point out that the traditional estimationmethodis not optimal if thecoefficientfunctions possess different degrees of smoothness. They propose a two-stepmethodto attenuate the drawback. To apply the two-stepmethod, one needs to identify whichcoefficientfunction is the smoothest one and undersmooth all other functions in the first stage, In this paper we proposeanalternativeapproach which assigns each smooth function a differentbandwidthand we choose the bandwidths simultaneously. Simulation results show that our proposedmethodcompare well with existing methods. (C) 2017 Elsevier B.V. All rights reserved.
ELSEVIER SCIENCE SA, PO BOX 564, 1001 LAUSANNE, SWITZERLAND
研究方向:Business & Economics
Web of Science 类别:Economics
