ldstatsHD (version 1.0.1)

aicAndbicLambdaSelection: AIC/BIC regularization parameter selection

Description

aicAndbicLambdaSelection is a function designed to select the regularization parameter in graphical models. It selects the graph with smallest AIC or BIC coefficients.

Usage

aicAndbicLambdaSelection(obj,  y, criterion = c("AIC", "BIC", "eBIC"), phi=1)

Arguments

obj

an object of class huge or camel.tiger.

y

original \(n \times p\) data set.

criterion

coefficients and optimal lambdas to be stored: to select from "AIC", "BIC" or "eBIC".

phi

weight used in the eBIC criterion (see reference).

Value

An object of class lambdaSelection containing the following components:

opt.lambda

optimal lambdas for AIC, BIC and eBIC.

crit.coef

coefficients for each lambda given the criterion AIC, BIC and eBIC.

criterion

with value defined by argument criterion.

References

Caballe, A., N. Bochkina, and C. Mayer (2016). Selection of the Regularization Parameter in Graphical Models using network charactaristics. eprint arXiv:1509.05326, 1-25.

Chen, J. and Z. Chen (2008). Extended Bayesian information criteria for model selection with large model spaces. Biometrika 95(3), 759-771.

See Also

lambdaSelection for other lambda selection approaches.

Examples

# NOT RUN {
# example to use aicAndBic function
EX1         <- pcorSimulator(nobs = 50, nclusters = 3, nnodesxcluster = c(40,30,30), 
                             pattern = "powerLaw")
y           <- EX1$y
Lambda.SEQ  <- seq(.35, 0.70, length.out = 40)
out3        <- huge(y, method = "glasso", lambda = Lambda.SEQ, cov.output = TRUE)
AIC.COEF    <- aicAndbicLambdaSelection(out3, y = y)
print(AIC.COEF)

 
 
# }