Estimating Hedonic Price Models by Stein Variance Double k-class estimators

In this paper the Stein variance Double k-class estimator is utilized to address the omitted variable issue in hedonic price modelling. If important housing attributes are excluded from the model, the estimated implicit prices of housing attributes would drift away from the true parameter value. Real estate researchers and practitioners are not completely ignorant about the parameters to be estimated. Experience and expertise usually provide them with tacit understanding of the likely range in which the true parameter value belongs. Under this scenario the subjective knowledge about the parameter value could be incorporated as non-sample information in the hedonic price model. The Stein variance Double k-class estimator is a biased estimator of linear regression coefficient with improved mean squared error terms. Theoretical and empirical research suggests that this methodology can effectively improve the precision of standard linear regression coefficient estimates and in-sample prediction within certain range of parameter space. The empirical evidence coincides with previous findings. Compared with traditional OLS approach, the Stein variance Double k-class estimators have more accurate predictive mean squared error terms and more precise parameter estimates.