ACL Training Materials Unit C: Analysing data characteristics 主要參考資料來源: KPMG ACL課程講義資料 Price. Water. House. Cooper ACL課程講義資料
Situations when we would use data analysis • ACL commands that allow us to perform data analysis: • Data validity commands; • Analysis of numeric fields and values; • Analysis of non-numeric fields and values.
Analysing data characteristics Numeric fields
Analysing data characteristics (數值) • Count • Total • Statistics • Profile • Stratify
Count • Count - counts the number of records that meet the specified condition;
Total • See if you can derive the total value of the invoices (ie voucher type ‘IN’) that were billed before 1997 • 可驗證資料之完整性
Total
Total Pricewaterhouse. Coopers
Statistics and Profile (統計分析) • Statistics - returns a statistical summary of one or more numeric fields; • Profile - returns similar but slightly less detailed information; • Run the statistics command on the value field.
Statistics and Profile (統計分析) Total value of receivables $468, 880. 69 Average value $865. 81 Positive values 609 records with value of $527, 277. 55 Negative values (probably credit notes) 161 records with value of $(58, 396. 86)
Analysing data characteristics Non-numeric fields Pw. C
Analysing data characteristics (非數值) • Classify • Age • Summarise
Analysing data characteristics • Using these commands we could: • Group data into subsets and return relevant totals for these subsets: --> Group data in a payroll file into departments and return the total payroll value for each department; • Perform ageing of records with date fields: --> Age a receivables file by intervals of 30 days to determine bad and doubtful debts.
Classify
Age • Performs ageing of records based upon a selected date field; • 可自行設定Cut - Off 的 日期,且可自訂區間
Age
Summarize Summarise • The order that the Summarise On fields are selected is very important