Imputation of Housing Rents for Owners Using Models With Heckman Correction

  • Beat Hulliger University of Applied Sciences Northwestern Switzerland
  • Gordon Wiegand University of Applied Sciences Northwestern Switzerland
Keywords: Selection bias, Statistics on Income and Living Conditions, survey data, imputed rents

Abstract

The direct income of owners and tenants of dwellings is not comparable since the owners have a hidden income from the investment in their dwelling. This hidden income is considered a part of the disposable income of owners. It may be predicted with the help of a linear model of the rent. Since such a model must be developed and estimated for tenants with observed market rents a selection bias may occur. The selection bias can be minimised through a Heckman correction. The paper applies the Heckman correction to data from the Swiss Statistics on Income and Living Conditions. The Heckman method is adapted to the survey context, the modeling process including the choice of covariates is explained and the effect of the prediction using the model is discussed.

Author Biographies

Beat Hulliger, University of Applied Sciences Northwestern Switzerland
P
Gordon Wiegand, University of Applied Sciences Northwestern Switzerland
Scientific Collaborator at School of Business FHNW
Published
2012-07-11
How to Cite
Hulliger, B., & Wiegand, G. (2012). Imputation of Housing Rents for Owners Using Models With Heckman Correction. Survey Research Methods, 6(2), 95-103. https://doi.org/10.18148/srm/2012.v6i2.5028
Section
Articles

Most read articles by the same author(s)