Abstract:

In financial accounting, data about the economic activities of a given firm is collected and then summarized and reported in the form of financial statements. Auditing, on the other hand, is the independent verification of the fairness of these financial statements. An item in an audit sample produces two pieces of information: the book (recorded) amount and the audited (correct) amount. The difference between the two is called the error amount. The book amounts are treated as values of a random variable whose distribution is a mixture of the distributions of the correct amount and the true amount contaminated by error. The mixing coefficient is equal to the proportion of the items with non-zero errors amounts. Below we consider a problem of testing appropriately formulated statistical hypotheses about admissibility of the total or the mean accounting errors. Hypotheses can be verified by the likelihood ratio test. The parameters of likelihood function can be estimated using the EM algorithm.