Evaluating errors in a classical variables sample
Concept Information
After you have performed your audit procedures on a set of sampled data you can use Analytics to:
- project the sample audit value to the entire account
- project any misstatements you found to the entire account
- calculate upper and lower limits on estimated total audit value, and estimated total misstatement amount
Even if you found no errors, you still use the evaluation feature to calculate the basic allowance for sampling risk. In this situation, you must use the mean-per-unit estimation type.
Follow the correct process for classical variables sampling
Evaluating errors is the final stage in the classical variables sampling process. You must have performed the previous stages before you can evaluate errors.
For more information, see:
Which estimation type should I use?
The CVS Evaluate stage gives you the option of four different estimation types (evaluation methods):
- Mean-per-unit
- Difference
- Ratio separate
- Ratio combined
The estimation type that you should use depends on the nature of the data: the sample book values, the sample audit values, and the relation between them.
Guidelines
The guidelines below help you select an estimation type.
Tip
If you want to compare the results produced by different estimation types, you can select All in the Estimation Type drop-down list. Selecting All includes all estimation types in the evaluation output.
Estimation type | Presence of misstatements | Size of misstatements | Sign of book values | Comparison of strata ratios |
---|---|---|---|---|
Mean-per-unit |
No misstatements, or very few misstatements The only valid estimation type if there are no misstatements, or very few misstatements, in the audited sample population. |
n/a | n/a | n/a |
Difference |
Misstatements required Requires a number of misstatements in the audited sample population. For example, 5% or more of the samples contain misstatements. |
Misstatements are non-proportional More suitable when misstatements are non-proportional: the size of a misstatement is not related to the size of the associated book value. In other words, small and large book values can have either small or large misstatements. |
n/a | n/a |
Ratio Separate |
Misstatements are proportional More suitable when misstatements are proportional: the size of a misstatement is related to the size of the associated book value. In other words, small book values have small misstatements, and large book values have large misstatements. |
Book values have the same sign All sample book values must have the same sign: either all positive, or all negative. |
Ratios vary More suitable when the ratio of average sample audit value to average sample book value varies widely between strata. |
|
Ratio Combined |
Ratios are consistent More suitable when the ratio of average sample audit value to average sample book value is relatively consistent between strata. |
Import the updated sample table
Import the updated sample table to Analytics from Excel, or from whatever external application you used to add audit values.
The table must contain:
- a book value field the original recorded book values that existed when you drew the sample
- an audit value field the audited values, updated where necessary based on the results of your analysis
For more information, see Import Microsoft Excel data.
Evaluate the results of the sample analysis
Note
Do not include the thousands separator, or the percentage sign, when you specify values. These characters prevent the command from running, or cause errors.
- Open the updated sample table that you just imported.
- Select Sampling > Classical Variables Sampling (CVS) > Evaluate.
Note
The menu option is disabled if a table is not open.
The CVS Evaluate dialog box opens. If you are using the output results of the CVS Prepare and the CVS Sample stages as input for the evaluation stage, most of the fields are prefilled with the required values.
If a number of the prefilled values are missing, you can:
- rerun the CVSSAMPLE command from the log to regenerate the values
- use the CVSEVALUATE command generated during the CVS Sample stage, if you saved it
Note
If you use the saved CVSEVALUATE command, you need to update the name of the audit value field, and possibly the evaluation type.
For more information, see CVSEVALUATE command.
- On the Main tab, select one of the following options from the Estimation Type drop-down list:
- MPU
- Difference
- Ratio Separate
- Ratio Combined
- All
Note
The options are explained in detail above.
- If you are not using prefilled values, or you want to adjust one or more values, enter or update the required values:
- Confidence Level (%)
- Number of Expected Errors
- Book Value
- Audit Value
- Precision Limits
- Top certainty stratum cutoff (Cutoff, Count, Value)
- Bottom certainty stratum cutoff (Cutoff, Count, Value)
- Strata Boundaries
- Population (Count, Value)
Note
The input values are explained in detail below.
Caution
Normally you should not change any of the prefilled values. Changing prefilled values can negate the statistical validity of the evaluation process.
- On the Output tab:
- In the To panel, select one of the following:
- Screen displays the results in the Analytics display area
Tip
You can click any linked result value in the display area to drill down to the associated record or records in the source table.
- File saves or appends the results to a text file
The file is saved outside Analytics.
- Screen displays the results in the Analytics display area
- If you selected File as the output type, do one of the following:
- Enter a file name in the Name text box.
- Click Name and enter the file name, or select an existing file in the Save or Save File As dialog box to overwrite or append to the file.
If Analytics prefills a file name, you can accept the prefilled name, or change it.
You can also specify an absolute or relative file path, or navigate to a different folder, to save or append the file in a location other than the project location. For example: C:\Results\Output.txt or Results\Output.txt.
Note
ASCII Text File or Unicode Text file (depending on which edition of Analytics you are using) is the only option for File Type.
- In the To panel, select one of the following:
- Click OK.
The CVS Evaluate output results are displayed, or saved to a file.
Note
The output results are explained in detail below.
For additional information about interpreting output results, see Judging whether the invoice records as a whole are fairly stated.
CVS Evaluate dialog box inputs and results
The tables below provide detailed information about the input values in the CVS Evaluate dialog box, and the output results.
Main tab – input values
Input values – CVS Evaluate dialog box |
Description |
---|---|
Estimation Type |
The estimation type (evaluation method) to use. |
Confidence Level (%) |
Your desired confidence level that the resulting sample is representative of the entire population. For example, entering 95 means that you want to be confident that 95% of the time the sample will in fact be representative. Confidence is the complement of “sampling risk”. A 95% confidence level is the same as a 5% sampling risk. |
Number of Expected Errors |
The minimum number of errors you expected in the sample. This value is not used in the CVS Evaluate calculation. Instead, it is used to trigger a notification if the actual number of errors you found in the sample is less than the Number of Expected Errors. If actual errors are fewer than Number of Expected Errors, the only evaluation method available is mean-per-unit. |
Book Value | The numeric field in the sample table containing the recorded book values. |
Audit Value | The numeric field in the sample table containing the audit values. |
Precision Limits |
The type of precision limit to use. For more information, see Preparing a classical variables sample. |
Top certainty stratum cutoff (Cutoff, Count, Value) |
The top certainty stratum cutoff value that was used in the CVS process, the number of records in the top certainty stratum, and their total value. |
Bottom certainty stratum cutoff (Cutoff, Count, Value) |
The bottom certainty stratum cutoff value that was used in the CVS process, the number of records in the bottom certainty stratum, and their total value. |
Strata Boundaries |
The boundary values that were used for stratifying the data set. |
Population (Count, Value) |
The number of records in each source table stratum, and the total value for each stratum. |
Output results
Output results – CVS Evaluate | Description |
---|---|
Evaluation Method | The estimation type you selected. |
Confidence Level | The confidence level that you specified as an input. |
Point Estimate |
A statistical projection of the most likely audited value of the entire data set in the source table. The Point Estimate is the midpoint of an estimated range. |
Precision |
A statistical projection of the amount by which the Point Estimate could vary. The Point Estimate plus or minus the Precision forms the upper and lower limits of the range. |
Estimated Total Audited Value |
A visual presentation of the Estimated Total Audited Value range. How the range works
|
Estimated Total Error |
A visual presentation of the Estimated Total Error range. How the Error range is calculated
How the range works
|