The Classical Linear Regression Model with one Incomplete Binary Variable

We present three di erent methods based on the conditional mean im putation when binary explanatory variables are incomplete Apart from the single imputation and multiple imputation especially the so called pi imputation is presented as a new procedure Seven procedures are com pared in a simulation experiment when missing data are con ned to one independent binary variable complete case analysis zero order regres sion categorical zero order regression pi imputation single imputation multiple imputation modi ed rst order regression After a brief theo retical description of the simulation experiment MSE ratio variance and bias are used to illustrate di erences within and between the approaches