Statistical Estimation of Incremental Cost from Accounting Data
Litigation assignments often require the computation of a plaintiff's damages. Damages typically reflect the incremental revenue less the incremental cost the plaintiff would have realized but for the defendant's alleged conduct. The incremental cost is typically the cost of producing and selling incremental units of output.
Experts sometimes use statistical regression analysis of accounting data to estimate the incremental cost.1 Properly applied, regression analysis can provide an accurate and reliable estimate of incremental cost along with an assessment of the accuracy of the estimate. Improperly applied, however, regression analysis can produce a biased and misleading estimate of incremental cost. This chapter covers both basic and advanced issues arising in the calculation of incremental cost for such damages cases using statistical regression analysis.
One can assume that the total cost incurred in producing a product most often comprises—or behaves as if it comprises—two kinds of cost: a fixed cost unaffected by the quantity produced and a variable cost that changes in proportion to the quantity produced.2 We write this assumption algebraically in terms of the following linear equation: