resp.check: Preliminary response diagnostics

In SemiParSampleSel: Semi-Parametric Sample Selection Modelling with Continuous or Discrete Response

Description

It takes takes the response vector and produces a histogram and QQ-plot. For discrete responses normalized and randomized QQ-plots are produced.

Usage

resp.check(y, margin = "N", bd = 1)

Arguments

y	reponse vector
margin	margin of interest. Is the response distributed according to the margin? The default is normal outcome.
bd	Binomial denominator - maximum value the response can take. This argument is only relevant in the case of BI, BB, ZIBI, ZABI, ZIBB and ZABB.

Author(s)

Maintainer: Giampiero Marra [e-mail]

Examples

library(SemiParSampleSel)

# Generating normal response with n = 2000
ys <- rnorm(2000, mean=3, sd=1)
resp.check(ys, margin = "N")

## Not run: 

# Generating gamma response with n = 2000
ys <- rgamma(2000, shape=2, rate=3)
resp.check(ys, margin = "G")

# Generating Poisson response with n = 2000
ys <- rPO(2000, mu=3)
resp.check(ys, margin = "P")

# Generating negative binomial response with n = 2000
ys <- rNBI(2000, mu=3, sigma=1)
resp.check(ys, margin = "NB")


## End(Not run)

SemiParSampleSel documentation built on May 2, 2019, 6:35 a.m.