drawLMmixed: Draws samples from the mixed LM model

In LMest: Generalized Latent Markov Models

Draws samples from the mixed LM model

Description

Function that draws samples from the mixed LM model with specific parameters.

Usage

drawLMmixed(la, Piv, Pi, Psi, n, TT, est = NULL,
            format = c("long","matrices"), seed = NULL)

Arguments

la	vector of mass probabilities for the first latent variable
Piv	matrix of initial probabilities of the latent Markov chain (k2 x k1)
Pi	set of transition matrices (k2 x k2 x k1)
Psi	array of conditional response probabitlies (mb x k2 x r)
n	sample size
TT	number of time occasions
est	object of class LMmixed (LMmixed-class)
format	character string indicating the format of final responses matrix
seed	an integer value with the random number generator state

Value

Y	matrix of response configurations unit by unit
S	matrix of distinct response configurations
yv	corresponding vector of frequencies
la	vector of mass probabilities for the first latent variable
Piv	matrix of initial probabilities of the latent Markov chain (k2 x k1)
Pi	set of transition matrices (k2 x k2 x k1)
Psi	array of conditional response probabitlies (mb x k2 x r)
n	sample size
TT	number of time occasions
est	object of class LMmixed (LMmixed-class)

Author(s)

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

Examples

# draw a sample for 1000 units and only one response variable and 5 time occasions
k1 <- 2
k2 <- 3
la <- rep(1/k1, k1)
Piv <- matrix(1/k2, k2, k1)
Pi <- array(0, c(k2, k2, k1))
Pi[,,1] <- diag(k2)
Pi[,,2] <- 1/k2
Psi <- cbind(c(0.6,0.3,0.1), c(0.1,0.3,0.6), c(0.3,0.6,0.1))
out <- drawLMmixed(la, Piv, Pi, Psi, n = 1000, TT = 5)


## Not run: 
# Example based on criminal data

data(data_criminal_sim)
data_criminal_sim = data.frame(data_criminal_sim)
# Estimate mixed LM model for females

responsesFormula <- lmestFormula(data = data_criminal_sim,
                                  response = "y")$responsesFormula
est <- lmestMixed(responsesFormula = responsesFormula,
                  index = c("id","time"),
                  k1 = 2,
                  k2 = 2,
                  data = data_criminal_sim[data_criminal_sim$sex == 2,])

out <- drawLMmixed(est = est, n = 100,  seed = 4321)


## End(Not run)

LMest documentation built on Aug. 27, 2023, 5:06 p.m.