SUBSTANTIVE AUDIT TESTS OF DETAILS AND THE ESTIMATION OF RARE ERRORS BY DOUBLE SAMPLING WITH PROBABILITY PROPORTIONAL TO SIZE (STATISTICS).
Item
-
Title
-
SUBSTANTIVE AUDIT TESTS OF DETAILS AND THE ESTIMATION OF RARE ERRORS BY DOUBLE SAMPLING WITH PROBABILITY PROPORTIONAL TO SIZE (STATISTICS).
-
Identifier
-
AAI8515631
-
identifier
-
8515631
-
Creator
-
HITZIG, NEAL B.
-
Contributor
-
Martin Benis
-
Date
-
1985
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Business Administration, Accounting
-
Abstract
-
This study is concerned with a problem that is commonly encountered in audits of financial statement data--using random sampling to detect and estimate infrequently occurring overstatement in an accounting population.;An action-oriented framework is developed and integrated with a set of sampling procedures, the objective of which is to provide an auditor with a reliable basis for making audit decisions concerning the amount of error that may exist in a population of accounting transactions or balances.;The principal statistical procedure developed in this study is a parametric method for obtaining interval estimates of the amount of error in a population under double sampling with probability proportional to size. The parametric procedure involves the augmentation of E. C. Molina's finding of the relationship between the Poisson distribution and the incomplete gamma function. The double sampling procedure is based on C. Stein's discovery of a test whose power is independent of the variance of the population.;The methods developed in this study are tested by computer-assisted analysis, which involves enumeration of the sample space in single and double sampling. Forty-two error conditions are analyzed under each of two levels of alpha risk (5 percent and 2 percent), varying amount of error (in three levels, from 50 percent to 150 percent of the specified maximum tolerable error) and richness of error condition (including binomial, trinomial and quadrinomial distributions).;The results of the analysis demonstrate that the proposed methodology comprises an effective audit test procedure whose estimation intervals are robust under nearly all of the error conditions to which the procedures were subjected.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
-
Program
-
Business
