effect.hetprob function

Get average marginal effects for heterokedastic binary models

Obtain the average marginal effects from hetprob class model.

## S3 method for class 'hetprob'
effect(object, vcov = NULL, digits = max(3, getOption("digits") - 2), ...)

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

## S3 method for class 'effect.hetprob'
print(x, ...)

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

Arguments

• object: an object of class hetprob and effect.hetprob for summary and print method.
• vcov: an estimate of the asymptotic variance-covariance matrix of the parameters for a hetprob object.
• digits: the number of digits.
• ...: further arguments.Ignored.
• x: an object of class effect.hetprob.

Returns

An object of class effect.heprob.

Details

This function allows to obtain the average marginal effects (not the marginal effects at the mean). The standard errors are computed using Delta Method.

Examples

# Data
library("AER")
data("PSID1976")
PSID1976$lfp  <- as.numeric(PSID1976$participation == "yes")
PSID1976$kids <- with(PSID1976, factor((youngkids + oldkids) > 0,
                                      levels = c(FALSE, TRUE), 
                                      labels = c("no", "yes")))
                                      PSID1976$finc <-  PSID1976$fincome / 10000
                                 
# Average marginal effects for heteroskedastic Probit model
labor_het <- hetprob(lfp ~  age + I(age^2) + finc + education + factor(kids) | 
                            factor(kids) + finc,              
                     data = PSID1976,                        
                     link = "probit")
eff_labor_het <- effect(labor_het)
summary(eff_labor_het)

Author(s)

Mauricio Sarrias.