A Collection of Techniques Correcting for Sample Selection Bias

[Up] [Top]

Documentation for package ‘sambia’ version 0.1.0

Help Pages

costing Predicting outcomes using Costing.
genSample Generate synthetic observations using inverse-probability weights
IPbag Predicting outcomes using non-parametric Inverse-Probability bagging
ipOversampling Plain replication of each observation by inverse-probability weights
rejSamp Rejection Sampling is a method used in sambia's function 'costing' (Krautenbacher et al, 2017).
smoteMod smoteMod is a modified version of the 'synthetic minority oversampling technique to generate new data.
smoteNew smoteNew is a necessary function that modifies the SMOTE algorithm.
synthIPbag Train a classifier via synthetic observations using inverse-probability weights