Changes for BAS 1.3.0 July 15, 2016 - vignetted added at last! - added crdible intervals for fitted and predicted values 'confint.pred.bas' - added credible intervals for coefficients 'confint.coef.bas' function - deprecated use of type to specify estimator in fitted.bas and replaced with 'estimator' - added standard errors for fitted values and predicted values in 'predict' - streamlined cv.summary.bas funbtion Changes for BAS 1.2.2 June 29, 2016 - added option to find "Best Predictive Model" or "BPM" for fitted.bas - added local Empirical Bayes prior and fixed g-prior for bas.glm - bug fix in plot.bas that appears with Sweave - bug fix in coef.bma when there is just one predictor - added diagnostic plot for MCMC - added truncated power prior as in Yang, Wainwright & Jordan (2016) Changes for BAS 1.2.1 April 16, 2016 - bug fix for method="MCMC" with truncated prior distributions where MH ratio was incorrect allowing models with 0 probability to be sampled. - fixed error in Zellner-Siow prior (ZS-null) when n=p+1 or saturated model where log marginal likelihood should be 0 Changes for BAS 1.2.0 April 11, 2016 - removed unsafe code where Rbestmarg (input) was being overwritten in .Call which would end up in corruption of the constant pool of the byte-code (Thanks to Tomas Kalibera for catching this!) - fixed issue with dimensions for use with Simple Linear Regression Changes for BAS 1.1.0 March 31, 2016 - New! added truncated Beta-Binomial prior and truncated Poisson (works only with MCMC currently) - deprecated method = "AMCMC" and issue warning message Probability model - improved code for finding fitted values under the Median - Changed S3 method for plot and image to use "bas" rather than 'bma' to avoid name conflicts with other packages Changes for BAS 1.09 - added weights for linear models - switched LINPACK calls in bayesreg to LAPACK finally should be faster - fixed bug in intercept calculation for glms - fixed inclusion probabilities to be a vector in the global EB methods for linear models Changes for BAS 1.08 - added intrinsic prior for GLMs - fixed problems for linear models for p > n and R2 not correct Changes for BAS 1.07 - added phi1 function from Gordy (1998) confluent hypergeometric function of two variables also known as one of the Horn hypergeometric functions or Humbert's phi1 - added Jeffrey's prior on g - added the general tCCH prior and special cases of the hyper-g/n. - TODO check shrinkage functions for all Changes for BAS 1.06 - new improved Laplace approximation for hypergeometric1F1 - added class basglm for predict - predict function now handles glm output - added dataframe option for newdata in predict.bas and predict.basglm - renamed coefficients in output to be 'mle' in bas.lm to be consistent across lm and glm versions so that predict methods can handle both cases. (This may lead to errors in other external code that expects object$ols or object$coefficients) - fixed bug with initprobs that did not include an intercept for bas.lm Changes for BAS 1.05 - added thinning option for MCMC method for bas.lm - returned posterior expected shrinkage for bas.glm - added option for initprobs = "marg-eplogp" for using marginal SLR models to create starting probabilities or order variables especially for p > n case - added standalone function for hypergeometric1F1 using Cephes library and a Laplace aproximation -Added class "BAS" so that predict and fitted functions (S3 methods) are not masked by functions in the BVS package: to do modify the rest of the S3 methods. Changes for BAS 1.04 - added bas.glm for model averaging/section using mixture of g-priors for GLMs. Currently limited to Logistic Regression - added Poisson family for glm.fit Changes for BAS 1.0 - cleaned up MCMC method code Changes for BAS 0.93 - removed internal print statements in bayesglm.c - Bug fixes in AMCMC algorithm Changes for BAS 0.92 - fixed glm-fit.R so that hyperparameter for BIC is numeric Changes for BAS 0.91 - added new AMCMC algorithm Changes for BAS 0.91 - bug fix in bayes.glm CHANGES for BAS 0.90 - added C routines for fitting glms CHANGES for BAS 0.85 - fixed problem with duplicate models if n.models was > 2^(p-1) by restricting n.models - save original X as part of object so that fitted.bma gives the correct fitted values (broken in version 0.80) CHANGES for BAS 0.80 - Added hypergeometric2F1 function that is callable by R - centered X's in bas.lm so that the intercept has the correct shrinkage - changed predict.bma to center newdata using the mean(X) - Added new Adaptive MCMC option (method = "AMCMC") (this is not stable at this point) CHANGES for BAS 0.7 -Allowed pruning of model tree to eliminate rejected models CHANGES for BAS 0.6 - Added MCMC option to create starting values for BAS (method = "MCMC+BAS") CHANGES for BAS 0.5 -Cleaned up all .Call routines so that all objects are duplicated or allocated within code CHANGES for BAS 0.45 - fixed ch2inv that prevented building on Windows in bayes glm_fit CHANGES for BAS 0.4 - fixed fortran calls to use F77_NAME macro - changed allocation of objects for .Call to prevent some objects from being overwritten. (still need to replace more) CHANGES for BAS 0.3 - fixed EB.global function to include prior probabilities on models - fixed update function CHANGES for BAS 0.2 - fixed predict.bma to allow newdata to be a matrix or vector with the column of ones for the intercept optionally included. - fixed help file for predict - added modelprior argument to bas.lm so that users may now use the beta-binomial prior distribution on model size in addition to the default uniform distribution - added functions uniform(), beta-binomial() and Bernoulli() to create model prior objects - added a vector of user specified initial probabilities as an option for argument initprobs in bas.lm and removed the separate argument user.prob