gm: Indicate Generalized Multivariate Model

In gmfamm: Generalized Multivariate Functional Additive Models

Indicate Generalized Multivariate Model

Description

This function is used in the formula call of a generalized multivariate functional additive mixed model to supply the information of the outcome and factor variables to bamlss.

Usage

gm(y, outcome, ...)

Arguments

y	Name of variable in data set which contains the values of the longitudinal outcome.
outcome	Name of variable in data set which is the factor variable indicating which outcome the value is from. Note that only the ordering not the factor levels are used in the estimation process.
...	Additional arguments not used at the moment.

Value

Matrix combining y and outcomes of class 'matrix' and 'gm'.

Examples

set.seed(123)
# Number of subjects
n <- 10

# Number of observations
ni <- 3

# Covariate vector
x <- rep(rnorm(n), each = ni)
t <- rep(c(0, 0.5, 1), times = n)

# Additive predictor
eta_1 <- t + 0.5*x
eta_2 <- t + 0.5*x

# Outcomes
y1 <- rnorm(n*ni, eta_1, 0.3)
y2 <- rbinom(n*ni, 1, 1/(1 + exp(-eta_2)))

# Data format
dat <- data.frame(
   id = factor(rep(seq_len(n), each = ni)),
   y = c(y1, y2),
   dim = factor(rep(c(1, 2), each = n*ni)),
   t = t,
   x = x,
   fpc = 1
)

# Specify formula
f <- list(
  gm(y, dim) ~ t + x,
  sigma1 ~ 1,
  mu2 ~ t + x,
  Lambda ~ -1 + s(id, by = fpc, bs = "re")
)

gmfamm documentation built on June 22, 2024, 10:35 a.m.