SMNlmec.sim {SMNlmec}R Documentation

Generating Censored UNC, DEC, CAR errors with Mixed Effects, for normal, student's-t and slash distribution.

Description

Generating Censored UNC, DEC, CAR errors with Mixed Effects, for normal, student's-t and slash distribution.

Usage

SMNlmec.sim(
  m,
  x,
  z,
  tt,
  nj,
  beta,
  sigma2,
  D,
  phi,
  struc = "UNC",
  typeModel = "Normal",
  p.cens = 0.1,
  n.cens = NULL,
  cens_type = "right",
  nu_set = NULL
)

Arguments

m

Number of individuals.

x

Design matrix of the fixed effects of order N x p, corresponding to vector of fixed effects.

z

Design matrix of the random effects of orderN x d, corresponding to vector of random effects.

tt

Vector N x 1 with the time the measurements were made, where N is the total number of measurements for all individuals.

nj

Vector m x 1 with the number of measurements of each individual, where m is the total number of individuals.

beta

Vector of values fixed effects.

sigma2

Values of the scalar of the variance matrix.

D

Variance matrix of the random effects of order d x d.

phi

Vector of parameter in the DEC and CAR structure. NULL for UNC, c(phi_1,phi_2) for DEC and c(phi_1,1) for CAR.

struc

Structure for the simulated data. Available options are UNC, DEC and CAR.

typeModel

Distribution of the simulated data. Available options are Normal, Student and Slash.

p.cens

Percentage of censored measurements in the responses. The default value is 0.1.

n.cens

Number of censored measurements in the responses. The default value is NULL.

cens_type

The direction of cesoring. Available options are left and right.

nu_set

degrees of freedom for student's-t or slash simulated data. The default value is NULL.

Value

return list:

cc

Vector of censoring indicators.

y_cc

Vector of responses censoring.

Examples

p.cens <- 0.1
m <- 50
D <- matrix(c(0.049,0.001,0.001,0.002),2,2)
sigma2_set <- 0.15
beta <- c(-2.83,-0.18)
nu <- 2
phi <- c(0.6,2)
nj <- rep(6,m)
tt <- rep(1:6,length(nj))
X1 <- rep(1,sum(nj))
X2 <- tt
x <- as.matrix(cbind(X1,X2))
Z1 <- rep(1,sum(nj))
Z2 <- tt
z <- as.matrix(cbind(Z1,Z2))

ID_sim <- rep(0,length(tt))
ID_log <- 0
for(i in 1:m) {
  for(j in 1:nj[i]) {
    ID_sim[ID_log + j] <- i
  }
  ID_log <- ID_log + nj[i]
}

Slash_DEC_sim <- SMNlmec.sim(m = m,x = x,z = z,tt = tt,nj = nj,beta = beta,
                      sigma2 = sigma2_set,D = D,phi= phi,struc ="DEC",
                      typeModel="Slash",p.cens = p.cens,n.cens = NULL,
                      cens_type="right",nu_set=nu)

head(Slash_DEC_sim$cc)
sum(Slash_DEC_sim$cc)/length(Slash_DEC_sim$cc)

head(Slash_DEC_sim$y_cc)


y_com <- as.numeric(Slash_DEC_sim$y_cc)
rho_com <- as.numeric(Slash_DEC_sim$cc)
tem <- tt

Slash_DEC_est <- SMNlmec.est(ID = ID_sim, x_set = x, z_set = z,
                         tt = tem, y_complete = y_com,
                         censor_vector = rho_com, dist = "Slash",
                         struc = "DEC", direction = "right",
                         thin_num = 1, chains_num = 1, iter_num = 3000,
                         burn_percen = 0.1, seed_set = 9955,
                         adapt_delta_set = 0.8)

SMNlmec.summary(Slash_DEC_est)



[Package SMNlmec version 0.1.0 Index]