confint.EMSS: Getting Confidence Intervals for Parameters of EM type Sample...

In SangkyuStat/EMSS: Some EM-Type Estimation Methods for the Heckman Selection Model

Description

confint method for a class "EMSS".

Usage

## S3 method for class 'EMSS'
confint(object, parm, level = 0.95, ...)

Arguments

object	an object of class "EMSS" made by the function EMSS.
parm	not used, but exists because of the compatibility.
level	a numeric value between 0 and 1 for controlling the significance level of confidence interval; default value is 0.95.
...	not used, but exists because of the compatibility.

Examples

# examples continued from EMSS
data(Smoke, package = "EMSS")
ex1 <- EMSS(response = cigs_intervals ~ educ,
           selection = smoker ~ educ + age,
           data = Smoke)
confint(ex1)

data(Smoke, package = "EMSS")
ex2 <- EMSS(response = cigs_intervals ~ educ,
           selection =  smoker ~ educ + age,
           data = Smoke, method="ECMnr")
confint(ex2)

## example using random numbers with exclusion restriction

N <- 1000
errps <- mvtnorm::rmvnorm(N,c(0,0),matrix(c(1,0.5,0.5,1),2,2) )
xs <- runif(N)
ys <- xs+errps[,1]>0
xo <- runif(N)
yo <- (xo+errps[,2])*(ys>0)

ex3 <- EMSS(response = yo ~ xo,
           selection = ys ~ xs,
           initial.param = c(rep(0,4), 0.3, 0.6), method="ECMnr")
confint(ex3)

SangkyuStat/EMSS documentation built on Jan. 1, 2022, 7:19 p.m.