CensoredNormal: Create a Censored Normal Distribution

In crch: Censored Regression with Conditional Heteroscedasticity

Create a Censored Normal Distribution

Description

Class and methods for left-, right-, and interval-censored normal distributions using the workflow from the distributions3 package.

Usage

CensoredNormal(mu = 0, sigma = 1, left = -Inf, right = Inf)

Arguments

mu	numeric. The location parameter of the underlying uncensored normal distribution, typically written \mu in textbooks. Can be any real number, defaults to 0.
sigma	numeric. The scale parameter (standard deviation) of the underlying uncensored normal distribution, typically written \sigma in textbooks. Can be any positive number, defaults to 1.
left	numeric. The left censoring point. Can be any real number, defaults to -Inf (uncensored). If set to a finite value, the distribution has a point mass at left whose probability corresponds to the cumulative probability function of the uncensored normal distribution at this point.
right	numeric. The right censoring point. Can be any real number, defaults to Inf (uncensored). If set to a finite value, the distribution has a point mass at right whose probability corresponds to 1 minus the cumulative probability function of the uncensored normal distribution at this point.

Details

The constructor function CensoredNormal sets up a distribution object, representing the censored normal probability distribution by the corresponding parameters: the latent mean mu = \mu and latent standard deviation sigma = \sigma (i.e., the parameters of the underlying uncensored normal variable), the left censoring point (with -Inf corresponding to uncensored), and the right censoring point (with Inf corresponding to uncensored).

The censored normal distribution has probability density function (PDF) f(x):

\Phi((left - \mu)/\sigma)	if x \le left
1 - \Phi((right - \mu)/\sigma)	if x \ge right
\phi((x - \mu)/\sigma)/\sigma	otherwise

where \Phi and \phi are the cumulative distribution function and probability density function of the standard normal distribution respectively.

All parameters can also be vectors, so that it is possible to define a vector of censored normal distributions with potentially different parameters. All parameters need to have the same length or must be scalars (i.e., of length 1) which are then recycled to the length of the other parameters.

For the CensoredNormal distribution objects there is a wide range of standard methods available to the generics provided in the distributions3 package: pdf and log_pdf for the (log-)density (PDF), cdf for the probability from the cumulative distribution function (CDF), quantile for quantiles, random for simulating random variables, crps for the continuous ranked probability score (CRPS), and support for the support interval (minimum and maximum). Internally, these methods rely on the usual d/p/q/r functions provided for the censored normal distributions in the crch package, see dcnorm, and the crps_cnorm function from the scoringRules package. The methods is_discrete and is_continuous can be used to query whether the distributions are discrete on the entire support (always FALSE) or continuous on the entire support (only TRUE if there is no censoring, i.e., if both left and right are infinite).

See the examples below for an illustration of the workflow for the class and methods.

Value

A CensoredNormal distribution object.

See Also

dcnorm, Normal, TruncatedNormal, CensoredLogistic, CensoredStudentsT

Examples

## package and random seed
library("distributions3")
set.seed(6020)

## three censored normal distributions:
## - uncensored standard normal
## - left-censored at zero (Tobit) with latent mu = 1 and sigma = 1
## - interval-censored in [0, 5] with latent mu = 1 and sigma = 2
X <- CensoredNormal(
  mu    = c(   0,   1, 1),
  sigma = c(   1,   1, 2),
  left  = c(-Inf,   0, 0),
  right = c( Inf, Inf, 5)
)

X


## compute mean and variance of the censored distribution
mean(X)
variance(X)
## higher moments (skewness, kurtosis) are not implemented yet

## support interval (minimum and maximum)
support(X)

## simulate random variables
random(X, 5)

## histograms of 1,000 simulated observations
x <- random(X, 1000)
hist(x[1, ], main = "uncensored")
hist(x[2, ], main = "left-censored at zero")
hist(x[3, ], main = "interval-censored in [0, 5]")

## probability density function (PDF) and log-density (or log-likelihood)
x <- c(0, 0, 1)
pdf(X, x)
pdf(X, x, log = TRUE)
log_pdf(X, x)

## cumulative distribution function (CDF)
cdf(X, x)

## quantiles
quantile(X, 0.5)

## cdf() and quantile() are inverses (except at censoring points)
cdf(X, quantile(X, 0.5))
quantile(X, cdf(X, 1))

## all methods above can either be applied elementwise or for
## all combinations of X and x, if length(X) = length(x),
## also the result can be assured to be a matrix via drop = FALSE
p <- c(0.05, 0.5, 0.95)
quantile(X, p, elementwise = FALSE)
quantile(X, p, elementwise = TRUE)
quantile(X, p, elementwise = TRUE, drop = FALSE)

## compare theoretical and empirical mean from 1,000 simulated observations
cbind(
  "theoretical" = mean(X),
  "empirical" = rowMeans(random(X, 1000))
)

## evaluate continuous ranked probability score (CRPS) using scoringRules
library("scoringRules")
crps(X, x)

crch documentation built on Sept. 30, 2024, 9:22 a.m.