论文信息 - A New Algorithm in Bayesian Model Averaging in Regression Models

A New Algorithm in Bayesian Model Averaging in Regression Models

We propose a new iterative algorithm, namely the model walking algorithm, to modify the widely used Occam’s window method in Bayesian model averaging procedure. It is verified, by simulation, that in the regression models, the proposed algorithm is much more efficient in terms of computing time and the selected candidate models. Moreover, it is not sensitive to the initial models.

[1] G. A. Barnard,et al. New Methods of Quality Control , 1963 .

[2] D. Madigan,et al. Bayesian Model Averaging for Linear Regression Models , 1997 .

[3] Keming Yu,et al. Bayesian Mode Regression , 2012, 1208.0579.

[4] J. York,et al. Bayesian Graphical Models for Discrete Data , 1995 .

[5] D. Madigan,et al. Model Selection and Accounting for Model Uncertainty in Graphical Models Using Occam's Window , 1994 .

[6] J. Brian Gray,et al. Introduction to Linear Regression Analysis , 2002, Technometrics.

[7] David Draper,et al. Assessment and Propagation of Model Uncertainty , 2011 .