主  題：Estimation for bivariate quantile varying coefficient model

內(nèi)容簡(jiǎn)介：We propose a bivariate quantile regression method for the bivariate varying  coefficient model through a directional approach. The varying coefficients  are approximated by the B-spline basis and an L2-type penalty is imposed to achieve desired smoothness. We develop a multistage estimation procedure  based on the Propagation-Separation (PS) approach to borrow information from nearby directions. The PS method is capable of handling the computational complexity raised by simultaneously considering multiple directions to efficiently estimate varying coefficients while guaranteeing certain smoothness along directions. We reformulate the optimization problem and solve it by the Alternating Direction Method of Multipliers (ADMM), which is implemented using R while the core is written in C to
 speed it up. Simulation studies are conducted to confirm the finite sample performance of our proposed method. A real data on Diffusion Tensor Imaging (DTI) properties from a clinical study on neurodevelopment is analyzed. Joint work with Haoxu Shu, Qianchuan Chad He, Giseon Heo, John
Gilmore, and Hongtu Zhu.

報(bào)告人：Linglong Kong   博導(dǎo)

時(shí)  間：2016-12-23    09:00

地  點(diǎn)：競(jìng)慧東樓305

舉辦單位：理學(xué)院 統(tǒng)計(jì)學(xué)與大數(shù)據(jù)研究院 科研部