Most modeling and analysis of audit populations has been based on surprisingly little formal study of actual empirical evidence. The single major empirical investigation to date has been the pioneering work of Neter and Loebbecke [1975; 1977]. This study examined the behavior of many statistical procedures commonly used in auditing, based on simulated audit populations constructed by extrapolating characteristics of errors found in audits of four actual populations. Neter and Loebbecke made a thorough examination of the four audit populations and reported on both the shapes of the book value distributions and error characteristics including error rate, magnitude, and changes in both rate and magnitude as a function of book value. In spite of their limited number, these four populations represented considerable variety in terms of type of account, degree of skewness, error rate, and balance of errors between underand overstatements. The effective size of the empirical data base was greatly expanded by the authors' careful control of these error characteristics in creating simulated populations with varying error rates. A major conclusion of the study was that a number of widely used statistical procedures, particularly those adopted from the sample survey literature, can be seriously misleading in the auditing context. This conclusion concurs with that of Kaplan [1973], who based his analysis on

John G. Ramage, Abba M. Krieger, Leslie L. Spero, [Discussion of An Empirical Study of Error Characteristics in Audit Populations]: A Reply, Journal of Accounting Research, Vol. 17, Studies on Auditing-Selections from the "Research Opportunities in Auditing" Program (1979), pp. 111-113

I. Income and consumption, 455.—II. Optimal consumption plans subject to borrowing constraint, 457.—III. Dependence of consumption on income in the short run: a consumption function, 460.—IV. Summary, 462.

M. June Flanders, Elhanan Helpman; An Optimal Exchange Rate Peg in a World of General Floating, The Review of Economic Studies, Volume 46, Issue 3, 1 July

A substantial portion of the research in input-output analysis has been concerned with attempts to estimate more efficiently the structural parameters. As the coefficients have generally been estimated by nonstochastic techniques, efficiency in this context has implied acceptable estimates at reduced cost. It is possible to distinguish at least two directions of effort. One group of analysts has sought to develop and standardize sampling techniques that would bring the cost of estimating these models within reasonable limits (cf. Miernyk, 1970; Isard and Langford, 1971). Another group has concentrated their effort on non-survey and partial survey techniques-the most widely discussed in recent years being the RAS method developed in the Department of Applied Economics at Cambridge (cf. Bacharach, 1970) and applied for the first time at the regional level by Czamanski and Malizia (1969). In an article in this REVIEW, Gerking (1976a) has argued that acceptable coefficients can be estimated using standard regression techniques. Gerking recognizes the stochastic nature of sample estimates required for the construction of input-output models, and it is argued that a rigorous examination of the value of input-output coefficients is only possible in this context. Elsewhere (Gerking, 1976b; 1976c), he has applied similar methods to the problem of sample size selection and the problem of reconciling row and column coefficients in an input-output context. It is the purpose of this note to demonstrate that the application of stochastic techniques to the estimation of regional input-output coefficients encounters difficulties that must be recognized before more serious work is done in this area. Specifically, we will argue that (1) the nature of the distribution of stochastic disturbances has not been adequately explored, (2) the unique nature of regional input-output models makes application of stochastic techniques particularly difficult, if not impossible, (3) the estimator of major interest in Gerking's article produces parameter values that are not constrained to satisfy both input-output identities, and further, that empirical application of this estimator fails because the associated finitesample distribution does not possess moments.

A well-established feature of the labor market is that an inverse relation exists between the unemployment rate (u) and the vacancy rate (v). Hansen (1970) provides a clear exposition of the theoretical model that implies that the u-v relation will be negatively sloped and convex to the origin. The usefulness of the relation is that it provides a means of distinguishing between changes in deficient demand unemployment (indicated by movements along the curve) and changes in nondeficient demand unemployment (indicated by a shift of the curve). In the United States, Great Britain and Canada there has been an upward shift of the u-v relation since the mid-sixties, although the timing and the magnitude of the shifts have varied considerably.' As figure 1 illus-