摘要:To compute the Blinder-Oaxaca decomposition and associated standard errors a practitioner needs to be comfortable using vector and matrix software manipulations. This paper proposes a computational trick for producing these computations by running an artificial regression.