In AndrMenezes/unitModalReg: Unit Modal Parametric Regression Models

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

unitModalReg: A Collection of Parametric Modal Regressions Models for Bounded Data

The goal of unitModalReg is to provide tools for fitting parametric modal regression models for bounded response variables.

Installation

You can install the development version from GitHub with:

# install.packages("remotes")
remotes::install_github("AndrMenezes/unitModalReg")

Example

The packages follows the structure of glm objects. The main function is unitModalReg.

library(unitModalReg)

# Simulate data
n <- 500
betas <- c(beta0 = 2, beta1 = -1, beta2 = 0.6)
x1  <- runif(n)
x2  <- rbinom(n, size = 1, prob = 0.65)
eta <- betas[1] + betas[2] * x1 + betas[3] * x2
mu  <- exp(eta) / exp(1 + eta)
phi <- 4
y   <- rugamma(n = n, mu = mu, phi = phi)
data_sim <- data.frame(y = y, x1 = x1, x2 = factor(x2))

# Families of distributions
distr <- list(
  Be = "betaMode",
  UGz = "unitGompertz",
  UGa = "unitGamma",
  Kum = "Kumaraswamy"
)

# Fit the models
fits <- lapply(distr, function(m) {
  unitModalReg(y ~ x1 + x2, data = data_sim, family = m, link = "logit")
})

# Get the coefficients
vapply(fits, coef, numeric(4))

We also implemented the gof function to compare the models based on the information criterion (AIC, BIC and HQIC) and Voung test for non-nested models.

# Compare the models fitting
gof(lt = fits)

The currently methods implemented are

methods(class = "unitModalReg")

AndrMenezes/unitModalReg documentation built on March 12, 2021, 5:24 a.m.