Function reference • NeEDS4BigData

Big Data

Real world big data

Electric_consumption: Electric consumption data

One_million_songs: One million songs data

Skin_segmentation: Skin segmentation data

Generate Big Data

Simulating big data under different conditions

GenGLMdata(): Generate data for Generalised Linear Models

GenModelMissGLMdata(): Generate data for Generalised Linear Models under model misspecification scenario

Subsampling

Model based Subsampling Algorithms

AoptimalGauLMSub(): A-optimality criteria based subsampling under Gaussian Linear Models

LCCsampling(): Local case control sampling for logistic regression

ALoptimalGLMSub(): A- and L-optimality criteria based subsampling under Generalised Linear Models

LeverageSampling(): Basic and shrinkage leverage sampling for Generalised Linear Models

AoptimalMCGLMSub(): A-optimality criteria based subsampling under measurement constraints for Generalised Linear Models

Model Robust Subsampling Algorithm

Subsampling algorithms when a set of models is considered to describe the big data

modelRobustLinSub(): Model robust optimal subsampling for A- and L- optimality criteria under linear regression

modelRobustLogSub(): Model robust optimal subsampling for A- and L- optimality criteria under logistic regression

modelRobustPoiSub(): Model robust optimal subsampling for A- and L- optimality criteria under Poisson regression

Model Misspecified Subsampling

Subsampling algorithms under a potentially misspecified model

modelMissLinSub(): Subsampling under linear regression for a potentially misspecified model

modelMissLogSub(): Subsampling under logistic regression for a potentially misspecified model

modelMissPoiSub(): Subsampling under Poisson regression for a potentially misspecified model

Supporting Functions

Functions to plot the results from the outputs

plot_Beta(): Plotting model parameter outputs after subsampling

plot_AMSE(): Plotting AMSE outputs for the samples under model misspecification