Multinom {kDGLM}

Multinom outcome for kDGLM models

Description

Creates an outcome with Multinomial distribution with the chosen parameters.

Usage

Multinom(p, data, offset = as.matrix(data)^0, base.class = NULL)

Arguments

p

character: a vector with the name of the linear predictor associated with the probability of each category (except the base one, which is assumed to be the last).

data

vector: Values of the observed data.

offset

vector: The offset at each observation. Must have the same shape as data.

base.class

character or integer: The name or index of the base class. Default is to use the last column of data.

Details

For evaluating the posterior parameters, we use the method proposed in Alves et al. (2024).

For the details about the implementation see dos Santos et al. (2024).

Value

A object of the class dlm_distr

References

Mariane Branco Alves, Helio S. Migon, Raíra Marotta, Junior, Silvaneo Vieira dos Santos (2024). “k-parametric Dynamic Generalized Linear Models: a sequential approach via Information Geometry.” 2201.05387.

Junior, Silvaneo Vieira dos Santos, Mariane Branco Alves, Helio S. Migon (2024). “kDGLM: an R package for Bayesian analysis of Dynamic Generialized Linear Models.”

See Also

fit_model

Other auxiliary functions for a creating outcomes: Gamma(), Normal(), Poisson(), summary.dlm_distr()

Examples

structure <- (
  polynomial_block(p = 1, order = 2, D = 0.95) +
    harmonic_block(p = 1, period = 12, D = 0.975) +
    noise_block(p = 1, R1 = 0.1) +
    regression_block(p = chickenPox$date >= as.Date("2013-09-01"))
  # Vaccine was introduced in September of 2013
) * 4

outcome <- Multinom(p = structure$pred.names, data = chickenPox[, c(2, 3, 4, 6, 5)])
fitted.data <- fit_model(structure, chickenPox = outcome)
summary(fitted.data)
plot(fitted.data, plot.pkg = "base")