The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

BrokenAdaptiveRidge

Build Status codecov.io

Introduction

BrokenAdaptiveRidge is an R package for performing L_0-based regressions using Cyclops

Features

Examples

library(Cyclops)
library(BrokenAdaptiveRidge)

## data dimension
p <- 30    # number of covariates
n <- 200   # sample size

## logistic model parameters 
itcpt     <- 0.2 # intercept
true.beta <- c(1, 0, 0, -1, 1, rep(0, p - 5))

## simulate data from logistic model
set.seed(100)

x <- matrix(rnorm(p * n, mean = 0, sd = 1), ncol = p)
x <- ifelse(abs(x) > 1., 1, 0)
y <- rbinom(n, 1, 1 / (1 + exp(-itcpt - x%*%true.beta)))


# fit BAR model
cyclopsData <- createCyclopsData(y ~ x, modelType = "lr")
barPrior    <- createBarPrior(penalty = 0.1, exclude = c("(Intercept)"), 
                             initialRidgeVariance = 1) 

cyclopsFit <- fitCyclopsModel(cyclopsData,
                             prior = barPrior)
fit1 <- coef(cyclopsFit) 

# fit BAR using sparse-represented covariates
tmp <- apply(x, 1, function(x) which(x != 0))

y.df <- data.frame(rowId = 1:n, y = y)
x.df <- data.frame(rowId = rep(1:n, lengths(tmp)), covariateId = unlist(tmp), covariateValue = 1)

cyclopsData <- convertToCyclopsData(outcomes = y.df, covariates = x.df, modelType = "lr")
barPrior    <- createFastBarPrior(penalty = 0.1, exclude = c("(Intercept)"), 
                                 initialRidgeVariance = 1) 

fit2 <- coef(cyclopsFit) 

# fit BAR using cyclic algorithm
cyclopsData <- createCyclopsData(y ~ x, modelType = "lr")
barPrior    <- createFastBarPrior(penalty = 0.1, exclude = c("(Intercept)"), 
                             initialRidgeVariance = 1) 

cyclopsFit <- fitCyclopsModel(cyclopsData,
                             prior = barPrior)
fit3 <- coef(cyclopsFit) 

fit1
fit2
fit3

Technology

System Requirements

Requires R (version 3.2.0 or higher).

Dependencies

Getting Started

  1. On Windows, make sure RTools is installed.
  2. In R, use the following commands to download and install BrokenAdaptiveRidge:
install.packages("devtools")
library(devtools)
install.packages("ohdsi/Cyclops") 
install_github("ohdsi/BrokenAdaptiveRidge") 
  1. To perform a L_0-based Cyclops model fit, use the following commands in R:
library(BrokenAdaptiveRidge)
cyclopsData <- createCyclopsData(formula, modelType = "modelType") ## TODO: Update
barPrior    <- createBarPrior(penalty = lambda / 2, initialRidgeVariance = 2 / xi) 
cyclopsFit  <- fitCyclopsModel(cyclopsData, prior = barPrior)
coef(cyclopsFit) #Extract coefficients

Getting Involved

License

BrokenAdaptiveRidge is licensed under Apache License 2.0.

Development

BrokenAdaptiveRidge is being developed in R Studio.

Acknowledgements

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.