tensorMAM

A tensor estimation approach to multivariate additive models.

For a high-dimensional multivariate additive model (MAM) using B-splines, with or without aparsity assumptions, treating the coefficients as a third-order tensor and borrowing Tucker decomposition to reduce the number of parameters.
The multivariate sparse group lasso (mcp or scad) and the coordinate descent algorithm are used to estimate functions for sparsity situation.

Installation

#install Rtools 3.5 (http://cran.r-project.org/bin/windows/Rtools)
#install.packages("devtools")
#install.packages("Rcpp")
library(devtools)
install_github("xliusufe/tensorMAM")

Usage

• tensorMAM-manual ------------ Details of the usage of the package.

Examples

library(tensorMAM)
# Example 1
# The usage of function "mam()"
p <- 5
q <- 5
D2 <- matrix(runif(2*p*q, 0.7, 1), 2, p*q) 
mydata <- generateData(200, q, p, p, D2) 
fit <- mam(mydata$Y, mydata$X)
K <- fit$K
D3hat <- fit$Dnew
D2hat <- TransferModalUnfoldings(D3hat,3,2,p,K,q)
mu <- fit$mu

# Example 2
# The usage of function "mam_dr()"
fit_dr <- mam_dr(mydata$Y, mydata$X)
K <- fit_dr$K
D3hat <- fit_dr$Dnew
D2hat <- TransferModalUnfoldings(D3hat,3,2,p,K,q)	
mu <- fit_dr$mu
opt <- fit_dr$rk_opt	

# Example 3 
# The usage of function "mvrblockwise()"
n <- 200
q <- 5
s <- 3
p <- 100
B <- matrix(runif(q*s, 2,3), s)
X <- matrix(rnorm(n*p),n,p)
Y <- X[,1:s]%*%B + matrix(rnorm(n*q),n)
fit <- mvrblockwise(Y,X) #See details in the function "mvrblockwise"
fit$activeX
fit$Bhat
which(rowSums(fit$Bhat^2)>0)
fit$muhat

# Example 4
# The usage of function "mvrcolwise()"
n <- 200
q <- 5
s <- 3
p <- 100
B <- matrix(runif(q*s, 2,3), s)
X <- matrix(rnorm(n*p),n,p)
Y <- X[,1:s]%*%B + matrix(rnorm(n*q),n)
fit <- mvrcolwise(Y,X) #See details in the function "mvrcolwise"
fit$activeX
fit$Bhat
which(rowSums(fit$Bhat^2)>0)
fit$muhat

References

A tensor estimation approach to multivariate additive models. Manuscript.

Development

The R-package is developed by Xu Liu ([email protected]) and Xiangyong Tan.