Improving Hedonic Estimation with an Inequality Restricted Estimator

摘要: Economists commonly estimate the value of characteristics not traded in explicit markets by hedonic pricing. Unfortunately, these nonexplicitly often display a lack independent variation or multicollinearity. Often some prior information on is available from submarkets. This paper utilizes this type to circumvent multicollinearity problems pricing models using an inequality restricted Bayesian estimator. The authors perform Monte Carlo experiment and cross-validation analysis demonstrate superiority over ordinary least squares at many margins variety situations typically faced estimation. Copyright 1995 MIT Press.