compute.loglik: Compute log likelihood

In anujkhare/iregnet: Regularized interval regression

Description

Compute log-likelihood of iregnet predictions, used to compute the surrogate loss on the validation set in cross-validation (cv.iregnet).

Usage

compute.loglik(y.mat, pred.mean, pred.scale, family)

Arguments

y.mat	numeric matrix of output variables (n x 2)
pred.mean	numeric matrix of mean parameters (predicted values, n x p)
pred.scale	numeric matrix of scale parameters (n x p)
family	gaussian, logistic

Value

numeric matrix of log-likelihood values (n x p)

Author(s)

Toby Dylan Hocking

anujkhare/iregnet documentation built on Aug. 23, 2019, 8:24 p.m.