genGlasso: generalized graphical lasso
In gherardovarando/crossSectional: Estimate Dynamical Model from Cross Sectional Data

Solve the generalized graphical lasso problem with proximal gradient

1
2
3

genGlasso(Sigma, P = diag(nrow(Sigma)), G = -diag(nrow(Sigma)),
  lambda = 0.1, maxIter = 1000, eps = 1e-05, alpha = 0.5,
  beta = 0.5, trace = 0)

`Sigma`	the covariance matriz
`P`	initial precision matrix
`G`	Matrix in the generlaized penalization
`lambda`	penalization coefficient
`maxIter`	maximum number of iterations
`eps`	threshold for stopping criteria
`alpha`	param for line search
`beta`	param for line search

The estimated precision matrix

gherardovarando/crossSectional documentation built on July 7, 2019, 12:44 a.m.

gherardovarando/crossSectional index

Note that we can't provide technical support on individual packages. You should contact the package authors for that.

Description