The article presents information on content of general price level earnings. The criticism of the use of a binomial test in which the author compare the number of cases in which the General Price Level (GPL) correlation coefficients exceed the historical cost (HC) coefficients, are correct. The problem with the test is that the observations are not independent. The results clearly indicate that, in the tests that we conducted, GPL earnings outperfouued HC earnings. The authors present results for the difference between GPL correlations and HC correlations. Based on the results in this table, the conclusions of this paper are criticized. The test is performed for three earnings ratios-net income to common equity, net income to market value of common equity and net income plus depreciation to market value of common equity. In addition, the test is performed for three portfolio sizes-individual securities, portfolios of five securities and portfolios often securities. The results for two out of the three earnings definitions, reveal that GPL earnings outperform HC earnings at the five percent significance level.
Auditors are charged with the responsibility of evaluating internal control systems. Recent advances in decision support systems indicate that the speed, accuracy, and memory capacity of computers may be used to aid auditors in this task. To test this proposition, a computer-assisted method of designing, analyzing, and evaluating internal control systems, called The Internal Control Model or TICOM, was designed and implemented. The technical manual and the TICOM software are available from the authors. This paper presents the results and conclusions of that project. TICOM is a computer-based analytic tool that aids the auditor first to model the internal control system and then to query the model in order to aid the auditor in evaluating the internal control system. TICOM is based on concepts in artificial intelligence such as knowledge representation and graph simplification. It serves (1) to describe office information systems while emphasizing internal accounting controls and (2) as an aid in control evaluation. It was found that the advantages of TICOM over traditional evaluation methods are that (1) the evaluation can be more rigorous and exhaustive, (2) the documentation of the system can be more thorough because of automated completeness and consistency tests, and (3) the modeler may probe and test controls by using the query-processing portion of TICOM.