Environment Based Clustering for Interpretable Predictive Models in High Dimensional Data
Calculates cluster summaries
Generate True Response vector for Linear Simulation
Generate True Response vector for Non-Linear Simulation
Cluster similarity matrix
Get selected terms from an earth object
Function to generate heatmap
Cluster data using environmental exposure
Prepare data for regression routines
Generate linear response data and test and training sets for simulatio...
Generate non linear response and test and training sets for non-linear...
Fit MARS Models on Simulated Cluster Summaries
Fit Multivariate Adaptive Regression Splines on Simulated Data
Simulate Covariates With Exposure Dependent Correlations
Fit Penalized Regression Models on Simulated Cluster Summaries
Fit Penalized Regression Models on Simulated Data
Plot Heatmap of Cluster Summaries by Exposure Status
Calculate Fisher's Z Transformation for Correlations
Companion package to the paper: An analytic approach for interpretable predictive models in high dimensional data, in the presence of interactions with exposures. Bhatnagar, Yang, Khundrakpam, Evans, Blanchette, Bouchard, Greenwood (2017) <DOI:10.1101/102475>. This package includes an algorithm for clustering high dimensional data that can be affected by an environmental factor.
Useful links