Using Table 6-2 as a guide, compare and contrast predictive and prescriptive analytics. How might these be used in an audit? Or a continuous audit?

Table 6-2

Transcribed Image Text:

Example CAATS Analytic Type Example Audit Procedure Age analysis-groups balances by date Descriptive-summarizes activity Analysis of new accounts opened and employee bonuses by employee and location. or masters data based on certain attributes Sorting-identifies largest or smallest values Summary statistics-mean, median, min, max, count, sum Sampling-random and monetary unit Diagnostic–detects correlations and patterns of interest Zscore-outlier detection Analysis of new accounts reveals that Jane Doe has an unusual Benford's law-identifies number of new accounts opened transactions or users with non- for customers who have been typical activity based on the distribution of first digits inactive for more than 12 months. Drill-down-explores the details behind the values Exact and fuzzy matching-joins tables and identifies plausible relationships Sequence check-detects gaps in records and duplicates entries Stratification-groups data by categories Clustering-groups records by non- obvious similarities Regression-predicts specific dependent values based on independent variable inputs Predictive-identifies common Analysis of new accounts opened attributes or patterns that may be used to identify similar activity for customers who have been inactive for more than 12 months collects data that are common Classification-predicts a category for a record to new account opening such as account type, demographics, and employee incentives. Probability–uses a rank score to evaluate the strength of classification Sentiment analysis-evaluates text for positive or negative sentiment to predict positive or negative outcomes What-if analysis-decision support Prescriptive-recommends action based on previously observed Analysis determines procedures to follow when new accounts are systems opened for inactive customers, such as requiring approval. actions Applied statistics-predicts a specific outcome or class Artificial intelligence–uses observations of past actions to predict future actions for similar events