Penalized regression via the restricted bridge estimator

This article is concerned with the Bridge Regression, which a special family in penalized regression penalty function $\sum_{j=1}^{p}|\beta_j|^q$ $q>0$, linear model restrictions. The proposed restricted bridge (RBRIDGE) estimator simultaneously estimates parameters and selects important variables when prior information about are available either low dimensional or high case. Using local quadratic approximation, term can be approximated around initial values vector RBRIDGE enjoys closed-form expression solved $q>0$. Special cases of our proposal LASSO ($q=1$), RIDGE ($q=2$), Elastic Net ($1< q < 2$) estimators. We provide some theoretical properties under for case, whereas computational aspects given both cases. An extensive Monte Carlo simulation study conducted based on different pieces performance estiamtor compared competitive estimators as well ORACLE. also consider four real data examples analysis comparison sake. numerical results show that suggested outperforms outstandingly true near exact