mvdnorm

<p>matrix with each row as an observed vector of multivariate normal RVs</p>

mean

sigma

<p>whether log transformation should be used</p>

logd

Fit Gaussian hidden Markov (or semi-Markov) models with / without autoregressive coefficients and with / without regularization. The fitting algorithm for the hidden Markov model is illustrated by Rabiner (1989) <doi:10.1109/5.18626>. The shrinkage estimation on the covariance matrices is based on the method by Ledoit et al. (2004) <doi:10.1016/S0047-259X(03)00096-4>. The shrinkage estimation on the autoregressive coefficients uses the elastic net shrinkage detailed in Zou et al. (2005) <doi:10.1111/j.1467-9868.2005.00503.x>.

mvdnorm: multivariate normal density

Description

Usage

Arguments

Value