logicDT1.0.5 package

Identifying Interactions Between Binary Predictors

bestBoostingIter

Get the best number of boosting iterations

calcAUC

Fast computation of the AUC w.r.t. to the ROC

calcBrier

Calculate the Brier score

calcDev

Calculate the deviance

calcMis

Calculate the misclassification rate

calcMSE

Calculate the MSE

calcNCE

Calculate the normalized cross entropy

calcNRMSE

Calculate the NRMSE

cooling.schedule

Define the cooling schedule for simulated annealing

cv.prune

Optimal pruning via cross-validation

fit4plModel

Fitting 4pL models

fitLinearBoostingModel

Linear models based on boosted models

fitLinearLogicModel

Linear models based on logic terms

fitLinearModel

Fitting linear models

get.ideal.penalty

Tuning the LASSO regularization parameter

getDesignMatrix

Design matrix for the set of conjunctions

gxe.test.boosting

Gene-environment (GxE) interaction test based on boosted linear models

gxe.test

Gene-environment interaction test

importance.test.boosting

Term importance test based on boosted linear models

logicDT.bagging

Fitting bagged logicDT models

logicDT.boosting

Fitting boosted logicDT models

logicDT

Fitting logic decision trees

partial.predict

Partial prediction for boosted models

plot.logicDT

Plot a logic decision tree

plot.vim

Plot calculated VIMs

predict.4pl

Prediction for 4pL models

predict.linear.logic

Prediction for linear.logic models

predict.linear

Prediction for linear models

predict.logicDT

Prediction for logicDT models

prune.path

Pruning path of a logic decision tree

prune

Post-pruning using a fixed complexity penalty

refitTrees

Refit the logic decision trees

splitSNPs

Split biallelic SNPs into binary variables

tree.control

Control parameters for fitting decision trees

vim

Variable Importance Measures (VIMs)

A statistical learning method that tries to find the best set of predictors and interactions between predictors for modeling binary or quantitative response data in a decision tree. Several search algorithms and ensembling techniques are implemented allowing for finetuning the method to the specific problem. Interactions with quantitative covariables can be properly taken into account by fitting local regression models. Moreover, a variable importance measure for assessing marginal and interaction effects is provided. Implements the procedures proposed by Lau et al. (2024, <doi:10.1007/s10994-023-06488-6>).

  • Maintainer: Michael Lau
  • License: MIT + file LICENSE
  • Last published: 2024-09-23