• Adaptive-Lasso for Cox's proportional hazards model (based on penalized partial likelihood)

         The software, written in the R language, conducts variable selection and model fitting simultaneously in the proportional hazards model      

         (Cox 1972). The adaptive-lasso estimate is obtained using the penalized partial likelihood method with weighted L-1 penalty.

         The algorithm iteratively solves least square problems with weighted L-1 penalty via a modified shooting algorithm of Fu (1998).

         Written by Wenbin Lu and Hao Helen Zhang.
         The original paper is to appear at Biometrika.
         Download the R code here.

• Adaptive-Lasso for proportional odds model (based on penalized marginal likelihood)

         The software, written in the R language, conducts variable selection and model fitting simultaneously in the proportional odds model      

         (Bennett 1983). The adaptive-lasso estimate is obtained using the penalized marginal likelihood method with weighted L-1 penalty.

         The algorithm iteratively solves least square problems with weighted L-1 penalty via a modified shooting algorithm of Fu (1998).

         Written by Wenbin Lu and Hao Helen Zhang.
         The original paper is to appear at Statistics in Medicine.
         Download the R code here.

• The marginal likelihood based boosting algorithm for nonlinear transformation models 

         The software, written in the R language, fits the nonlinear transformation models using a smoothing spline based boosting learning algorithm

         (e.g. Buhlmann and Yu, 2003; Li and Luan, 2005). We first compute the log marginal likelihood function and its gradient using importance             

         sampling. Then we use the negative gradients as responses and fit the model using componentwise cubic smoothing spline as the base learner.

         Written by Wenbin Lu and Lexin Li.
         Download the R code here.