ob_decompose_calculate_vcov

The function calculate the covariance matrix for the decomposition terms
of the linear Oaxaca-Blinder decomposition assuming independence between
groups.

Implements the Oaxaca-Blinder decomposition method and generalizations of it that decompose differences in distributional statistics beyond the mean.
The function ob_decompose() decomposes differences in the mean outcome between two groups into one part explained by different covariates (composition effect) and into another part due to differences in the way covariates are linked to the outcome variable (structure effect). The function further divides the two effects into the contribution of each covariate and allows for weighted doubly robust decompositions. For distributional statistics beyond the mean, the function performs the recentered influence function (RIF) decomposition proposed by Firpo, Fortin, and Lemieux (2018).
The function dfl_decompose() divides differences in distributional statistics into an composition effect and a structure effect using inverse probability weighting as introduced by DiNardo, Fortin, and Lemieux (1996). The function also allows to sequentially decompose the composition effect into the contribution of single covariates.
References:
Firpo, Sergio, Nicole M. Fortin, and Thomas Lemieux. (2018) <doi:10.3390/econometrics6020028>. "Decomposing Wage Distributions Using Recentered Influence Function Regressions."
Fortin, Nicole M., Thomas Lemieux, and Sergio Firpo. (2011) <doi:10.3386/w16045>. "Decomposition Methods in Economics."
DiNardo, John, Nicole M. Fortin, and Thomas Lemieux. (1996) <doi:10.2307/2171954>. "Labor Market Institutions and the Distribution of Wages, 1973-1992: A Semiparametric Approach."
Oaxaca, Ronald. (1973) <doi:10.2307/2525981>. "Male-Female Wage Differentials in Urban Labor Markets."
Blinder, Alan S. (1973) <doi:10.2307/144855>. "Wage Discrimination: Reduced Form and Structural Estimates."

Samuel Meier

ddecompose

Detailed Distributional Decomposition

ob_decompose_calculate_vcov function

<dl><dt>beta0</dt>
<dd>vector of estimated coefficients of group 0</dd>
<dt>beta1</dt>
<dd>vector of estimated coefficients of group 1</dd>
<dt>X0</dt>
<dd><code>model.matrix</code> of group 0</dd>
<dt>X1</dt>
<dd><code>model.matrix</code> of group 1</dd>
<dt>weights0</dt>
<dd>vector of observation weights of group 0</dd>
<dt>weights1</dt>
<dd>vector of observation weights of group 1</dd>
<dt>Cov_beta0</dt>
<dd>estimated covariance matrix of coefficients of group 0</dd>
<dt>Cov_beta1</dt>
<dd>estimated covariance matrix of coefficients of group 1</dd>
<dt>reference_0</dt>
<dd>boolean: indicating if group 0 is the reference group and if its coefficients are used to compute the counterfactual mean.</dd></dl>

Arguments

Calculate covariance matrix for OB decomposition terms — ob_decompose_calculate_vcov

<dl>

<dt>beta0</dt>
<dd>vector of estimated coefficients of group 0</dd>


<dt>beta1</dt>
<dd>vector of estimated coefficients of group 1</dd>


<dt>X0</dt>
<dd><code>model.matrix</code> of group 0</dd>


<dt>X1</dt>
<dd><code>model.matrix</code> of group 1</dd>


<dt>weights0</dt>
<dd>vector of observation weights of group 0</dd>


<dt>weights1</dt>
<dd>vector of observation weights of group 1</dd>


<dt>Cov_beta0</dt>
<dd>estimated covariance matrix of coefficients of group 0</dd>


<dt>Cov_beta1</dt>
<dd>estimated covariance matrix of coefficients of group 1</dd>


<dt>reference_0</dt>
<dd>boolean: indicating if group 0 is the reference group and if its coefficients are used to compute the counterfactual mean.</dd>

</dl>

ob_decompose_calculate_vcov: Calculate covariance matrix for OB decomposition terms

Description

Usage

Arguments

References