Reviews the book "Accounting for Data Processing Costs," by Robert W. McGee.

This paper demonstrates that the treatment of self-services, their inclusion or exclusion, affects neither the decision relevance of the information made available by the reciprocal method of cost allocation nor the internal versus external acquisition decision made through the underlying technology matrix. Three propositions and proofs are provided, showing that there is a one-to-one transformation between the two treatments of self-services in each of the analyses.

The article presents a comment on the effect of chance variation on revenue and cost estimations for breakeven analysis. In traditional breakeven analysis, total revenue and total cost are represented by straight lines with one intersection which indicates the breakeven quantity output. The effect of quadratic revenue and cost curves on breakeven analysis was introduced. Two breakeven points for the quadratic model were derived by solving for the levels of output at which the regressed total revenue equals total cost. Regardless of whether a linear or non-linear breakeven model may be appropriate, however, the revenue and cost functions are often unknown. Givens illustrated the regression analysis, but did not consider the effect of regression forecasting errors. In order to resolve the above issue an attempt was made to demonstrate the determination of what he called "the chance variation" in estimated revenue and cost functions; and to illustrate how these chance variations affect the results of a breakeven analysis. The estimated cost-volume functions were subtracted from the estimated revenue-volume functions to derive an estimated profit-volume function.

There is often evidence of confusion between two forms of the constant percentage learning curve model: the cumulative average and the individual unit forms. Failure to distinguish between the two models can lead to their misuse and to potentially serious errors of estimation. A precise statement of the difference between the two clarifies the errors of misspecification. This note provides an analytical comparsion of the two models and addresses empirical estimation issues. Résumé. On a souvent la preuve de la confusion entre deux formules de modèle de la courbe d'apprentissage à pourcentage constant: celle de la moyenne cumulative et celle de l'unité individuelle. Le fait de ne pas distinguer les deux modèles peut conduire à une mauvaise utilisation et éventuellement à de sérieuses erreurs d'estimation. Un examen précis de la différence entre les deux peut clarifier les erreurs de spécification. Cet article présente une comparaison analytique des deux modèles et traite des questions d'estimation empirique.

Reviews the book "The Capital Expenditure," by Arthur V. Corr.