Partial factor modeling: : Predictor-dependent shrinkage for linear regression

• Autores: P. Richard Hahn, Carlos M. Carvalho, Sayan Mukherjee
• Localización: Journal of the American Statistical Association, ISSN 0162-1459, Vol. 108, Nº 503, 2013, págs. 999-1008
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • We develop a modified Gaussian factor model for the purpose of inducing predictor-dependent shrinkage for linear regression. The new model predicts well across a wide range of covariance structures, on real and simulated data. Furthermore, the new model facilitates variable selection in the case of correlated predictor variables, which often stymies other methods.