pwbart: Predicting new observations with a previously fitted BART.predict.wbart: Predict new observations with a fitted BART model.predict.pbart: Predict new observations with a fitted BART model.permute.vs: Permutation-based variable selection approach.pbart: Probit BART for binary responses with Normal latents.mixtwo: Generate data with correlated and mixed-type predictors.mixone: Generate data with independent and mixed-type predictors.medianInclusion.vs: Variable selection with DART.
mc.wbart: BART for continuous responses with parallel computation.mc.pwbart: Predicting new observations based on a previously fitted BART.mc.permute.vs: Permutation-based variable selection approach with parallel.mc.pbart: Probit BART for binary responses with parallel computation Function allPerms enumerates all possible permutations for the number of observations and the selected permutation scheme.mc.backward.vs: Backward selection with two filters (using parallel.mc.abc.vs: Variable selection with ABC Bayesian forest (using parallel.friedman: Generate data for an example of Friedman (1991).checkerboard: Generate data for an example of Zhu, Zeng and Kosorok (2015).bartModelMatrix: Create a matrix out of a vector or data frame.an object of class gossetbtpermute with the final BTm() model, selected variables, seeds (random numbers) used for permutations and deviances Author(s) Jonathan Steinke and Kau de Sousa. BartMixVs-package: Varibale Selection Using Bayesian Additive Regression Trees In turn, variable selection continuous as long as any real variable has stronger explanatory power for pairwise rankings than the random variables.abc.vs: Variable selection with ABC Bayesian forest Part of R Language Collective 24 I trained a random forest using caret + ranger.
0 kommentar(er)
