The Cost of Drilling for Oil and Gas: An Application of Constrained Robust Regression

Abstract The robust regression method of Huber (1973) is used to fit a model to the cost of drilling for petroleum. Because the model includes a categorical variable (well type), a linear constraint is imposed on the parameter estimates Because the model was fit to the logarithm of cost and because it will be used to make repeated predictions of cost, an adjustment that approximately unbiases the predictions is imposed. The numerical values of the estimates are discussed, and a comparison is made with ordinary least squares.