greg {pgraph} | R Documentation |
greg
calculate the regularized graphical model estimation using lasso, scad and adaptive lasso penalties. It report the results in the form of roc results for each method.
greg(z, A, eps = 1e-15, rholist = NULL, gamma = 0.5, trace = FALSE)
z |
n * p dimensional matrix |
A |
p * p true graph |
eps |
a tolerence level for thresholding |
rholist |
a sequence of penalty parameters |
gamma |
the adaptive lasso penalty parameter |
trace |
whether to trace to estimation process. |
a list.
roc.lasso |
roc results for lasso |
roc.scad |
roc results for scad |
roc.alasso |
roc results for adaptive lasso |
set.seed(0)
p = 20;
n = 300;
tmp=runif(p-1,1,3)
s=c(0,cumsum(tmp));
s1=matrix(s,p,p)
cov.mat.true=exp(-abs(s1-t(s1)))
prec.mat.true=solve(cov.mat.true);
a=matrix(rnorm(p*n),n,p)
data.sa=a%*%chol(cov.mat.true);
true.graph = outer(1:p,1:p,f<-function(x,y){(abs(x-y)==1)})
greg.fit = greg(data.sa, true.graph)
auc.lasso = sum(diff(greg.fit$roc.lasso[,1])*greg.fit$roc.lasso[-1,2])
auc.alasso = sum(diff(greg.fit$roc.alasso[,1])*greg.fit$roc.alasso[-1,2])
auc.scad = sum(diff(greg.fit$roc.scad[,1])*greg.fit$roc.scad[-1,2])
auc.lasso
auc.alasso
auc.scad