Abstract The article presents response of the author on comments made by scholar Peter C. Fishburn on his article "Theory Versus Practice in Risk Analysis: An Empirical Study." Fishburn has provided an excellent mathematical proof of the inconsistency between the von Neumann and Morgenstern expected utility model and the mean-standard deviation model used in my paper. In fact, his point can be made in much more intuitive terms. Also consider two investments with different means, but with the same standard deviations. The impact on g of the risk will be identical for the two investments, again violating the requirement, f < 0. However, I must point out that the general increasing function suggested by Professor Fishburn also violates the f requirement. Notice that Firm 2 chose "R" for investment E and "N" for investment L. Since Fishburn disdosed no inconsistencies between the actual decisions and those made by the general increasing model for Firm 2, the same pair of decisions must have been made by the general increasing model. These decisions call for attaching a larger risk discount to Investment L than to F. Yet L lies strictly above E on the function, so the f, requirement is violated.

Abstract This article presents a study on the theory versus practice in risk analysis in decision making in the U.S. There appears to be substantial conflict between the decision processes used by actual decision makers and existing utility theory. The conflict seems to center around the inability of classical utility theory to deal effectively with situations where one or more contingent outcomes for a project are lower than some critical amount. Existing theory, therefore, incorrectly models practice. Contributing factors in the apparent conflict may be internal inconsistency and a tendency decision makers have to be more averse to risk at the time of actual choice than their pre-decision statements would indicate.

Abstract The article is a comment on an article "Discounted Cash Flows, Price Level Adjustments and Expectations," by researcher, Harold Bierman Jr., published in the October 1971 issue of the journal "The Accounting Review." Bierman suggests that the book value of his $1,529 asset would be shown as $1,036 on a price-level adjusted position statement which is "too high a value since the present value of the asset is $901." This "finding" leads to his major conclusion. According to the author, such a statement is not consistent with the manner in which price-level adjustments are and should be made. In fact, price-level adjustments are simply translations from one unit of currency to another. The projection of cash flows is correct but it must be noted that this schedule is expressed in monetary amounts reflecting the change in price level between periods. A schedule can, and should, be re-cast into three separate schedules, each using a uniform value of currency. A correct statement of economic depreciation for the first period, has been provided.

Abstract During the preparation of a capital budgeting study the decision-maker frequently encounters a situation where the timing of an important economic event is subject to uncertainty. Most typical of this class of problem is one in which doubt exists as to the useful life of a proposed machine. The analyst is aware that the associated revenue stream will come to an end at some future time, but it is not often that this time can be specified exactly. Similar characteristics are present in the uncertain timing of a research and development project's pay-off period and in the time lag between an advertising expenditure and the resulting increase in revenue. Several different approaches are now taken in facing the problem of uncertain timing. One common method is to treat the decision-maker's estimate of the most likely time as though it were certain. This estimate may be derived through a careful subjective probability analysis or simply from an intuitive feeling. If, greater precision is needed for some special application, there are two alternatives to the laborious task of computing the probabilities by hand. Those who have access to a digital computer will find it well worth while to write or otherwise obtain a program to produce probabilities.

Abstract This article presents response of the author on the comments made by scholar C.G. Hoskins on the paper "Theory Versus Practice in Risk Analysis: An Empirical Study," that was published in the July 1974 of the periodical "The Accounting Review." The comment by Hoskins represents a significant advance in modeling actual decision processes. However, one must add that Hoskins' findings support rather than dispute the conclusion of the original study that there appears to be substantial conflict between the decision processes used by actual decision makers and existing utility theory. The author's intentions in writing this response are to demonstrate why this is so and to expand further on the Hoskins model. It was restated that the model in the original study was not designed to produce decisions, which closely paralleled actual decisions. Rather, the model was built in the shape of classical utility theory with parameters, which were computed from an explicit statement by the firms as to their attitude toward risk-taking. The principal conclusion of the original research was that actual decisions are inconsistent with such a model.

Abstract Reviews the book "The Function and Analysis of Capital Market Rates," by James C. Van Horne.