LMbasiccont-class: Class "LMbasiccont"

In LMest: Generalized Latent Markov Models

Class 'LMbasiccont'

Description

An S3 class object created by lmestCont function for the latent Markov (LM) model for continuous responses in long format.

Value

lk	maximum log-likelihood
piv	estimate of initial probability vector
Pi	estimate of transition probability matrices (k x k x TT)
Mu	estimate of conditional means of the response variables (r x k)
Si	estimate of var-cov matrix common to all states (r x r)
np	number of free parameters
k	optimal number of latent states
aic	value of the Akaike Information Criterion for model selection
bic	value of the Bayesian Information Criterion for model selection
lkv	log-likelihood trace at every step
n	number of observations in the data
TT	number of time occasions
modBasic	model on the transition probabilities: default 0 for time-heterogeneous transition matrices, 1 for time-homogeneous transition matrices, 2 for partial time homogeneity based on two transition matrices one from 2 to (TT-1) and the other for TT
sepiv	standard errors for the initial probabilities
sePi	standard errors for the transition probabilities
seMu	standard errors for the conditional means
seSi	standard errors for the var-cov matrix
sc	score vector
J	information matrix
Lk	vector containing the values of the log-likelihood of the LM model with each k (latent states)
Bic	vector containing the values of the BIC of the LM model with each k (latent states)
Aic	vector containing the values of the AIC of the LM model with each k (latent states)
V	array containing the posterior distribution of the latent states for each units and time occasion
Ul	matrix containing the predicted sequence of latent states by the local decoding method
Pmarg	matrix containing the marginal distribution of the latent states
call	command used to call the function
data	data frame given in input

Author(s)

Francesco Bartolucci, Silvia Pandolfi, Fulvia Pennoni, Alessio Farcomeni, Alessio Serafini

See Also

lmestCont

LMest documentation built on April 3, 2025, 10:26 p.m.