Robert N. Stavins and Adam B. Jaffe (1990), ‘Unintended Impacts of Public Investments on Private Decisions: The Depletion of Forested Wetlands’, American Economic Review, 80, pp. 337-52.

MA frequently used measure of dynamic performance is the root-mean-squared (RMS) error, but this measure suffers from the limitation that its magnitude is not standardized. An alternative is Theil’s inequality coefficient (Theil, 1961). The numerator of this statistic is the RMS error, and the scaling of its denominator ensures that values fall within the bounds of zero and unity, where zero indicates a perfect dynamic fit. In keeping with the ordering of most goodness-of-fit measures, the final comparative statistic shown in Table 2 is equal to one minus the Theil coefficient, so that a perfect fit is evidenced by a value of 1.0.