RCASPAR

RCASPAR implements a piecewise baseline hazard Cox regression model with an Lq-norm prior to perform variable selection and predict survival times from high-dimensional explanatory covariates in genomics and molecular biology datasets.

Key Features:

• Piecewise Baseline Hazard Cox Regression Model: Implements a Cox regression model with a piecewise baseline hazard to model complex time-to-event hazard structures.
• Lq-Norm Based Prior: Uses an Lq-norm prior for shrinkage and selection of regression coefficients to identify relevant covariates.
• High-Dimensional Data Handling: Targets analysis of datasets with numerous explanatory covariates typical of genomic and molecular biology studies.

Scientific Applications:

• Survival Analysis: Prediction and modeling of survival times and other time-to-event outcomes using high-dimensional covariates.
• Genomics and Molecular Biology: Analysis of high-throughput genomic and molecular datasets where variable selection and survival modeling are required.

Methodology:

Computational methods explicitly include a piecewise baseline hazard Cox regression model and an Lq-norm based prior for selection of regression coefficients applied to high-dimensional explanatory covariates.