Bayesian variable selection for linear regression with the κ-G priors

In this paper, we introduce a new methodology for Bayesian variable selection in linear regression that is independent of the traditional indicator method. A diagonal matrix $\mathbf{G}$ introduced to prior coefficient vector $\boldsymbol{\beta}$, with each $g_j$'s, bounded between $0$ and $1$, on serves as stabilizer corresponding $\beta_j$. Mathematically, promising has $g_j$ value close $0$, whereas an unpromising $1$. This property proven paper under orthogonality together other asymptotic properties. Computationally, sample path obtained through Metropolis-within-Gibbs sampling Also, give two simulations verify capability selection.