Inferring About the Average Value of Audit Errors from Sequential Ratio Tests.

Area of Science:

• Accounting research
• Statistical modeling
• Auditing methodology

Background:

• Estimating average audit errors is a significant challenge in economic research.
• Book amounts are modeled as a mixture of correct and error-contaminated values.
• The mixing coefficient indicates the proportion of items with non-zero errors.

Purpose of the Study:

• To determine the necessary sample size for statistical hypothesis testing concerning mean accounting errors.
• To provide a robust method for sample size estimation in auditing contexts.

Main Methods:

• Utilizing a mixture distribution model for book amounts.
• Employing the Sequential Probability Ratio Test (SPRT) for hypothesis testing.
• Applying the Monte Carlo method to estimate the average sample size.

Main Results:

• The study provides an estimation of the average sample size required for hypothesis testing on mean accounting errors.
• The methodology allows for sample size determination under conditions of error-contaminated data.

Conclusions:

• The developed approach offers a statistically sound method for sample size determination in accounting error analysis.
• This research contributes to improving the efficiency and reliability of audit procedures through better sample size planning.