Variable selection is an important topic in linear regression analysis especially in high-dimensional medical data sets, The Least Absolute Shrinkage and Selection Operator (LASSO) proposed by Tibshirani is a popular technique for model selection and estimation in linear regression model. The Least Angle Regression (LAR) procedure by Efron et al. (2004) provides a method for fast computation of LASSO solution in regression problems. L1 penalized estimation methods shrink the estimates of the regression coefficients towards zero relative to the maximum likelihood estimates. The purpose of this shrinkage is to prevent over fit arising due to either collinearity of the covariates or high-dimensionality. L1 penalty tends to result in many regression coefficients shrunk exactly to zero and a few other regression coefficients with comparatively little shrinkage. It is important to note that shrinkage methods are generally not invariant to the relative scaling of the covariates. Variable selection for LAD regression receives much attention in recent literature. In this paper it is proposed to study the variable selection for survival analysis using LASSO Vs LAD regression. Numerical illustrations are substantiated through real data example
