DirichletReg: DirichletRegModel – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

DirichletRegModel

Methods for the Class DirichletRegModel

Description

These are available methods for the results of Dirichlet regression models and objects of class DirichletRegModel. These methods contain functions for print and summary of the data, generate fitted values and predicting new values using predict. Various types of residuals are implemented and confint can be used to compute confidence intervals of the parameters. Furthermore logLik extracts the log-likelihood of the model and vcov extracts the covariance matrix of the parameter estimates.

Usage

## S3 method for class 'DirichletRegModel'
print(x, digits = max(3, getOption("digits") - 3), ...)

## S3 method for class 'DirichletRegModel'
summary(object, ...)

## S3 method for class 'DirichletRegModel'
fitted(object, mu = TRUE, alpha = FALSE, phi = FALSE, ...)

## S3 method for class 'DirichletRegModel'
predict(object, newdata, mu = TRUE, alpha = FALSE, phi = FALSE, ...)

## S3 method for class 'DirichletRegModel'
residuals(object, type = c("standardized", "composite", "raw"), ...)

## S3 method for class 'DirichletRegModel'
confint(object, parm, level, ..., type=c("all", "beta", "gamma"), exp = FALSE)

## S3 method for class 'DirichletRegConfint'
print(x, digits = 3, ...)

## S3 method for class 'DirichletRegModel'
logLik(object, ...)

## S3 method for class 'DirichletRegModel'
AIC(object, ..., k = 2)

## S3 method for class 'DirichletRegModel'
BIC(object, ...)

## S3 method for class 'DirichletRegModel'
nobs(object, ...)

## S3 method for class 'DirichletRegModel'
vcov(object, ...)

## S3 method for class 'DirichletRegModel'
update(object, formula., ..., evaluate = TRUE)

## S3 method for class 'DirichletRegModel'
drop1(object, scope, test = c("LRT", "none"), k = 2, sort = TRUE, ...)

Arguments

`x`	an object of class `DirichletRegModel`
`object`	an object of class `DirichletRegModel` or `DirichletRegConfint` for printing an object obtained by `confint.DirichletRegModel`
`alpha`	logical; returns alpha values
`mu`	logical; returns expected values
`phi`	logical; returns precision values
`type`	for `residuals`: defines the type of residuals to be computed `"standardized"` (i.e., Pearson), `"composite"`, or `"raw"` for `confint`: defines the type of parameter (`"all"`, `"beta"`, or `"gamma"`) for which confidence values are returned
`newdata`	a `data.frame` containing new observations
`k`	number for the weighting of parameters
`parm`	a vector containing names of the parameters to print
`level`	(a vector of) confidence level(s), defaults to `.95`
`exp`	logical; returns parameters in exponentiated form
`digits`	the number of digits in the output
`formula.`	the new formula to be updated, see `update.formula` and `update.Formula`
`evaluate`	if `FALSE` the updated call will be returned, but not evaluated
`scope`	defines the scope of variables to be dropped, see `drop1`
`test`	defines the type of test for `drop1`
`sort`	if `TRUE`, p-values will be sorted in decreasing order.
`...`	further arguments

Author(s)

Marco J. Maier

Examples

ALake <- ArcticLake
ALake$AL <- DR_data(ArcticLake[, 1:3])

mod1 <- DirichReg(AL ~ depth + I(depth^2) | depth, data = ALake, model="alternative")

update(mod1, . ~ . | . + I(depth^2), evaluate = FALSE)
mod1

drop1(mod1)   ### issues a caveat when used for the first time in an R session

summary(mod1)

head(fitted(mod1))

predict(mod1, newdata = data.frame("depth" = seq(10, 100, 10)))

head(residuals(mod1))

confint(mod1)
confint(mod1, exp = TRUE)

logLik(mod1)
round(vcov(mod1), 5)

DirichletReg

Dirichlet Regression in R

v0.7-0

GPL (>= 2)

Authors

Marco Johannes Maier [cre, aut] (<https://orcid.org/0000-0002-1715-7456>)

Initial release

2020-05-28