Calculates permutation-based importance measures for individual predictors and interactions within a logic regression tree in a logic forest.
pimp.import(fit, data, testdata, BSpred, pred, Xs, mtype)A list with the following components:
Vector of importance estimates for individual predictors.
Vector of importance estimates for interactions (pimps).
Matrix indicating which predictors (and NOT-ed predictors) are used in each interaction.
Vector of predictor IDs used in the tree.
Vector of predictor column indices corresponding to vec.Xvars.
Fitted logic regression tree object containing outcome, model type, and logic tree information.
In-bag sample (training data).
Out-of-bag sample (test data).
Number of predictors included in the interactions (includes NOT-ed variables).
Number of predictors in the model (used for constructing permuted matrices).
Matrix or data frame of 0/1 values representing all predictor variables.
Model type: "classification", "linear", "Cox-PH Time-to-Event", or "Exp. Time-to-Event".
Bethany J. Wolf wolfb@musc.edu
J. Madison Hyer madison.hyer@osumc.edu
This function calculates importance measures for each bootstrapped sample by comparing model fit between the original out-of-bag sample and a permuted out-of-bag sample. Model fit is evaluated using:
Misclassification rate for classification models,
Log2 mean squared error for linear regression,
Harrell's C-index for survival regression (Cox-PH or exponential time-to-event models).
Wolf BJ, Hill EG, Slate EH. Logic Forest: an ensemble classifier for discovering logical combinations of binary markers. Bioinformatics. 2010;26(17):2183–2189. tools:::Rd_expr_doi("10.1093/bioinformatics/btq354")
logforest