회계정보의 왜곡형태가 분석적 절차의 효과성에 미치는 영향

The auditor obtains and evaluates evidential matter regarding management's assertions embodied in the financial statements to form an opinion. The auditor uses analytical procedures, tests of details, or a combination of both as substantive audit procedures to validate a client's reported balance. But analytical procedures and tests of details differ in effectiveness and efficiency. Hence, the efficient combination of both is critical consideration to the auditor's survival in a competitive audit environment. Past studies show the superiority of regression model using monthly data over other methods of analytical procedures in terms of efficiency(measured by sample size) and effectiveness(measured by detection risk). Even though monthly data-based regression model provides promising results, the data reliability is questionable because the auditor does not audit monthly data in detail. A primary use of analytical procedures is to derive the minimal level of detailed tests needed to achieve the desired level of audit risk. However, significant problems may arise from using the regression-based analytical procedures if the model misleads the level of detailed tests needed. This study employs three independent variables to predict sales revenue and considers two types of errors. The results show that detection risks achieved from the correct data are well below the auditor's planned level of detection risk. But achieved detection risks from the data containing immaterial overstatement errors exceed those from the correct data and the auditor's planned level. Random errors appear to be less of a concern. The results suggest that the auditor should adjust immaterial overstatement errors left from the past audit or consider at least the causes of those errors, since random errors do not affect the detection risks as badly as the overstatement errors.