There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
Principal Investigator | |
Principal Investigator's Name: | Qiwei Wu |
Institution: | University of Missouri, Columbia |
Department: | Department of Statistics |
Country: | |
Proposed Analysis: | Variable selection is a commonly asked question and has been investigated extensively. In practice, a biomedical or clinical study usually involves two sets of covariates: low-dimensional demographic or environmental factors and high-dimensional biomarkers or gene expressions. Detecting important biomarkers while taking into account subject's demographic information is usually of interest and under such scenario, regression analysis with semi-parametric covariate effects is usually considered. This paper focuses on high-dimensional variable selection for Cox's model for interval-censored data with semiparametric covariate effects. Bernstein polynomials are used to approximate the nonparametric part of the covariate effects and a coordinate-wise optimization algorithm, which is able to accommodate most existing penalties, is proposed to simultaneously estimate the parameters and select variables in the parametric part. The numerical study in this paper shows that both the approximation for the non-parametric part and the variable selection for the parametric part perform well. |
Additional Investigators |