version 1.3

DESCRIPTION

@@ -1,9 +1,9 @@
 Package: glasso 
 Title:  Graphical lasso- estimation of Gaussian graphical models
-Version: 1.2
+Version: 1.3
 Author: Jerome Friedman, Trevor Hastie and Rob Tibshirani
 Description:  Graphical lasso
 Maintainer: Rob Tibshirani <tibs@stat.stanford.edu>
 License: GPL-2
 URL: http://www-stat.stanford.edu/~tibs/glasso
-Packaged: Wed Jun 18 23:50:37 2008; tibs
+Packaged: Thu Jan  8 07:40:34 2009; tibs

NAMESPACE

@@ -1,2 +1,3 @@
 export(glasso)
+export(glassopath)
 useDynLib(glasso) 

R/glasso.R

@@ -35,7 +35,6 @@ if(start.type=="warm"){
       ww=w.init
       xx=wi.init
   }
-
 
 itrace=1*trace
 ipen=1*(penalize.diagonal)

R/glassopath.R

@@ -0,0 +1,43 @@
+glassopath=function (s, rholist=NULL, thr = 1e-04, maxit = 10000, approx = FALSE
+, 
+                penalize.diagonal = TRUE,  w.init = NULL, 
+                wi.init = NULL, trace = FALSE) 
+{
+        n = nrow(s)
+if(is.null(rholist)){
+      rholist=seq(max(abs(s))/10,max(abs(s)),length=10)
+  }
+rholist=sort(rholist)
+                itrace = 1 * trace
+                ipen = 1 * (penalize.diagonal)
+                ia = 1 * approx
+                rho=xx=ww=matrix(0,n,n)
+                nrho=length(rholist)
+                beta=what=array(0,c(n,n,nrho))
+                jerrs=rep(0,nrho)
+                mode(rholist) = "single"
+                mode(nrho) = "integer"
+                mode(rho) = "single"
+                mode(s) = "single"
+                mode(ww) = "single"
+                mode(xx) = "single"
+                mode(n) = "integer"
+                mode(maxit) = "integer"
+                mode(ia) = "integer"
+                mode(itrace) = "integer"
+                mode(ipen) = "integer"
+                mode(thr) = "single"
+                mode(beta) = "single"
+                mode(what) = "single"
+                mode(jerrs) = "integer"
+
+        junk <- .Fortran("glassopath", beta=beta,what=what,jerrs=jerrs,rholist,nrho,n, s, rho, ia,        itrace, ipen,
+        thr, maxit = maxit, ww = ww, xx = xx, niter = integer(1), 
+        del = single(1), ierr = integer(1), PACKAGE="glasso")
+
+    xx = array(junk$beta,  c(n,n,nrho))
+    what = array(junk$what,  c(n,n,nrho))
+    return(list(w=what, wi = xx, 
+        approx = approx, rholist=rholist, errflag = junk$jerrs))
+}
+

man/glasso.Rd

@@ -26,7 +26,7 @@ Only needed when start="warm" is specified}
 Only needed when start="warm" is specified}
 \item{trace}{Flag for printing out information as iterations proceed.
 Default FALSE}
-}}
+}
 
 \details{
 Estimates a sparse inverse covariance matrix using a lasso (L1) penalty,
@@ -55,7 +55,7 @@ Jerome Friedman, Trevor Hastie and Robert Tibshirani (2007).
 Sparse inverse covariance estimation with the lasso.
 Biostatistics 2007. http://www-stat.stanford.edu/~tibs/ftp/graph.pdf
 
-Meinshausen, N. and  B\"{u}hlmann, P.(2006)
+Meinshausen, N. and  Buhlmann, P.(2006)
  High dimensional graphs
 and variable selection with the lasso.
 Annals of Statistics,34, p1436-1462.

man/glassopath.Rd

@@ -0,0 +1,76 @@
+\name{glassopath}
+\alias{glassopath}
+\title{Compute the Graphical lasso  along a path}
+\description{
+Estimates a sparse inverse covariance matrix using a lasso (L1) penalty, along a path
+of values for the regularization parameter
+}
+\usage{
+glassopath(s, rholist=NULL, thr=1.0e-4, maxit=1e4,  approx=FALSE, penalize.diagonal=TRUE, w.init=NULL,wi.init=NULL, trace=FALSE)
+}
+\arguments{
+\item{s}{Covariance matrix:p by p matrix (symmetric)}
+\item{rholist}{Vector of non-negative regularization parameters for the lasso.
+Should be  increasing from smallest to largest; actual path is computed from largest
+ to smallest value of rho).
+ If NULL, 10 values in a   (hopefully reasonable) range are used. Note that
+the same parameter rholist[j] is used for all entries of the inverse covariance matrix;
+different penalties for different entries are not allowed. }
+\item{thr}{Threshold for convergence. Default value is 1e-4.  Iterations stop when average absolute parameter change is less than thr * ave(abs(offdiag(s)))}
+\item{maxit}{Maximum number of iterations of outer loop. Default 10,000}
+\item{approx}{Approximation flag: if true, computes Meinhausen-Buhlmann(2006)
+approximation}
+\item{penalize.diagonal}{Should diagonal of inverse covariance be penalized?
+Dafault TRUE.}
+\item{w.init}{Optional starting values for estimated covariance matrix (p by p).
+Only needed when start="warm" is specified}
+\item{wi.init}{Optional starting values for estimated inverse covariance matrix (p by p)
+Only needed when start="warm" is specified}
+\item{trace}{Flag for printing out information as iterations proceed.
+Default FALSE}
+}
+
+\details{
+Estimates a sparse inverse covariance matrix using a lasso (L1) penalty, along a path of
+regularization paramaters,
+using the approach of Friedman, Hastie and Tibshirani (2007).
+The Meinhausen-Buhlmann (2006) approximation is also implemented. 
+The algorithm can also be used to estimate a graph with missing edges,
+by specifying which edges to omit in the zero argument, and setting rho=0.
+Or both fixed zeroes for some elements and regularization on the other elements
+can be specified.
+}
+\value{
+A list with components
+\item{w}{Estimated covariance matrices, an array  of dimension (nrow(s),ncol(n), length(rholist))}
+\item{wi}{Estimated inverse covariance matrix, an array  of dimension (nrow(s),ncol(n), length(rholist))}
+\item{approx}{Value of input argument approx}
+\item{rholist}{Values of regularization parameter used}
+\item{errflag}{values of error flag (0 means no memory allocation error)}
+}
+
+
+ \references{
+Jerome Friedman, Trevor Hastie and Robert Tibshirani (2007).
+Sparse inverse covariance estimation with the lasso.
+Biostatistics 2007. http://www-stat.stanford.edu/~tibs/ftp/graph.pdf
+
+Meinshausen, N. and  Buhlmann, P.(2006)
+ High dimensional graphs
+and variable selection with the lasso.
+Annals of Statistics,34, p1436-1462.
+}
+\examples{
+
+
+set.seed(100)
+
+x<-matrix(rnorm(50*20),ncol=20)
+s<- var(x)
+a<-glassopath(s)
+
+}
+\keyword{multivariate}
+\keyword{models}
+\keyword{graphs}
+