Function reference • cssr

All functions

clusterRepLasso(): Select features via the cluster representative lasso (Bühlmann et. al. 2013)

corFunction(): Absolute value of sample correlation between two vectors

createSubsamples(): Creates lists of subsamples for stability selection.

css(): Cluster Stability Selection

cssLasso(): Provided fitfun implementing the lasso

cssLoop(): Helper function run on each subsample

cssPredict(): Wrapper function to generate predictions from cluster stability selected model in one step

cssSelect(): Obtain a selected set of clusters and features using cluster stability selection

formCssDesign(): Create design matrix of cluster representatives from matrix of raw features using results of css function

formatClusters(): Formats clusters in standardized way, optionally estimating cluster prototypes

genClusteredData(): Generate randomly sampled data including noisy observations of latent variables

genClusteredDataWeighted(): Generate randomly sampled data including noisy observations of latent variables, where proxies differ in their relevance (noise level)

genClusteredDataWeightedRandom(): Generate randomly sampled data including noisy observations of latent variables, where proxies differ in their relevance (noise level)

genZmuY(): Generates Z, weak signal features in X, noise features in X, mu, and y from provided parameters

getAllClustWeights(): Calculate weights for each cluster member of all of the selected clusters.

getClustWeights(): Calculate weights for members of a cluster using selection proportions

getClusterSelMatrix(): Get cluster selection matrix

getClusterSelsFromGlmnet(): Extracts selected clusters and cluster prototypes from the glmnet lasso output

getCssDesign(): Obtain a design matrix of cluster representatives

getCssPreds(): Fit model and generate predictions from new data

getCssSelections(): Obtain a selected set of clusters and features

getLassoLambda(): Get lambda value for lasso

getModelSize(): Automated estimation of model size

getNoiseVar(): Get variance of noise to add to Z in order to yield proxies X with desired correlations with Z

getPrototypes(): Estimate prototypes from a list of clusters

getSelMatrix(): Generates matrix of selection indicators from stability selection.

getSelectedClusters(): From css output, obtain names of selected clusters and selection proportions, indices of all selected features, and weights of individual cluster members

getSelectedSets(): Converts a selected set from X_glmnet to selected sets and selected clusters from the original feature space of X.

getSelectionPrototypes(): Identify selection prototypes from selected clusters

getSubsamps(): Generate list of subsamples

getXglmnet(): Converts the provided design matrix to an appropriate format for either the protolasso or the cluster representative lasso.

identifyPrototype(): Estimate prototypes from a single cluster

print(<cssr>): Print cluster stability selection output

printCssDf(): Prepares a data.frame summarazing cluster stability selection output to print

processClusterLassoInputs(): Check the inputs to protolasso and clusterRepLasso, format clusters, and identify prototypes for each cluster

protolasso(): Select features via the protolasso (Reid and Tibshirani 2016)