heckitmfx_log: Marginal Effects for Heckit (Log-Tranformed Variable)

In santosglauber/heckitmfx: Marginal Effects for Two-Step Sample Selection Models (Heckit)

Description

Partial and total marginal effects for two-step sample selection models (Heckit) with log-transformed outcome variable. Partial marginal effects refer to marginal effects associated with changes in the probability of selection and with changes in the conditional outcome. Total marginal effects refer to marginal effects on the unconditional outcome.

Usage

heckitmfx_log(data, selection, outcome)

Arguments

data	The dataset
selection	The selection equation
outcome	The outcome equation

Value

A dataframe with estimated parameters, partial and total marginal effects

References

Hoffmann, R. & Kassouf, A. L. (2005) Deriving conditional and unconditional marginal effects in log earnings equations estimated by Heckman's procedure, Applied Economics, 37:11, 1303-1311, DOI: https://www.doi.org/10.1080/00036840500118614. (Equation 14).

Examples

# Example following Hoffmann & Kassouf (2005) specification (eqs. 1 & 2)
## Not run: heckitmfx_log(data, L ~ z1 + z2, log(g) ~ x1 + x2)

# Example using simulated data of household tourism expenditures (in logarithms)
data(tourexp)
selection <- "participation ~ income + education + health"
outcome <- "log(expenditure) ~ income + education + tripweather"
heckitmfx_log(tourexp, selection, outcome)

santosglauber/heckitmfx documentation built on March 4, 2022, 12:42 a.m.