Deriving Benefit Measures with Higher Precision: A Study of Economic Values of Air Quality

A calibration strategy using ridge regression to generate more precise estimates for a particular parameter in a model is proposed. Formulae to compute the proposed ridge estimates from standard ordinary least squares (OLS) results are provided. The strategy is applied to recomputing marginal effects of air pollution on property values for 30 housing studies. Results show that ridge estimates are superior to the OLS estimates under the mean squared error criterion. The same strategy can be used to reestimate key parameters of interest in other applications, such as price elasticities for demand forecasts or the value of a statistical life from hedonic wage regressions.