“…It also includes a graphical representation of the model space via an image plot. The flexibility and user-friendliness of the package is proved by the fact that various studies have used package BMS to perform Bayesian model averaging (among others, see Giannone, Lenza, and Reichlin 2011;Horváth 2011;Amini and Parmeter 2012;Babecký, Havránek, Matěju, Rusnák,Šmídková, and Vašíček 2013;Horváth 2013;Iršová and Havránek 2013;Feldkircher 2014;Feldkircher, Horváth, and Rusnák 2014;Horváth, Rusnák,Šmídková, and Zapal 2014). Future versions of BMS might include spatial regression models (Crespo Cuaresma and Feldkircher 2013;Crespo Cuaresma, Doppelhofer, and Feldkircher 2014), model averaging based on the predictive likelihood (Eklund and Karlsson 2007;Feldkircher 2012), so-called "heredity priors" to handle related predictors (Chipman 1996) and dilution priors that penalize models which contain highly collinear regressors (George 2010).…”