Performing monetary unit sampling
Concept Information
You can create a new table that contains a representative sample of the monetary data in the active table.
Monetary unit sampling is appropriate if you are interested in the total amount of monetary misstatement in a file.
Note
This procedure does not include filtering (IF statements) or scope parameters because applying these options compromises the validity of a sample.
Steps
Note
Detailed information appears after the steps. See Sample dialog box options.
Do not include the thousands separator, or the percentage sign, when you specify values. These characters prevent the command from running, or cause errors.
- In the Navigator, open the table you want to draw a sample from.
- Optional. If you intend to use the Random selection method, profile or generate statistics on the sample field.
- Select Sampling > Record/Monetary Unit Sampling > Sample.
- On the Main tab, select MUS.
- Select the field to sample from the Sample On dropdown list.
- In the Sample Parameters panel, specify a sample selection method:
- Fixed Interval
- Cell
- Random
Note
Do not use the random selection method with monetary unit sampling if you intend to use Analytics to evaluate any misstatement detected in the resulting sample.
Evaluating monetary unit samples requires that you use the fixed interval or cell selection methods.
- Enter the sample parameters for the selection method you chose:
Selection method Sample parameters Fixed Interval - Interval
- Start (optional)
- Cutoff (optional)
Cell - Interval
- Seed (optional)
- Cutoff (optional)
- Algorithm – leave Mersenne Twister selected
Random - Size
- Seed (optional)
- Population
- Algorithm – leave Mersenne Twister selected
Note
Sample parameters are explained in detail below.
- In the To text box, specify the name of the Analytics table that will contain the output results.
- On the More tab, select one
of the following:
RecordThe entire record is included in the output table.
Fields Only the selected fields are included in the output table.
- If you chose Fields, select the field(s) to include in the output table from the Extract Fields list.
- Optional. Select one or more of the following options:
- Subsample
- Report Selection Order
- No Repeats
Note
The options are explained below.
Subsample is available only if Fields output is selected.
Report Selection Order is available only if both the Random selection method and Fields output are selected.
- Click OK.
Sample dialog box options
The tables below provide detailed information about the options in the Sample dialog box.
Main tab
Options – Sample dialog box | Description |
---|---|
MUS
Record |
The sample type:
|
Sample On |
The numeric sample field.
|
Fixed Interval |
Specifies that the fixed interval method is used for sample selection. Samples are selected based on an interval value and a start number that you specify. For more information, see Fixed interval selection method. If you selected Fixed Interval enter the following values:
|
Cell |
Specifies that the cell method is used for sample selection. The data set is divided into multiple equal-sized cells or groups, and one sample is randomly selected from each cell. The interval value dictates the size of each cell. For more information, see Cell selection method. If you selected Cell enter the following values:
|
Random |
Specifies that the random method is used for sample selection. Samples are randomly selected from the entire data set. For more information, see Random selection method. If you selected Random enter the following values:
|
If |
Caution Do not create an IF statement or filter records in the course of sampling. Doing so compromises the validity of the sample. For more information, see Conditional sampling. |
To | The name and location of the output table.
Regardless of where you save the output table, it is added to the open project if it is not already in the project. If Analytics prefills a table name, you can accept the prefilled name, or change it. Note Analytics table names are limited to 64 alphanumeric characters, not including the .FIL extension. The name can include the underscore character ( _ ), but no other special characters, or any spaces. The name cannot start with a number. |
Local |
If you are connected to a server table, specifies where to save the output table.
|
Use output table | Specifies whether the Analytics table containing the output results opens automatically upon completion of the operation. |
More tab
Options – Sample dialog box | Description |
---|---|
Scope panel |
Caution Do not limit which records are processed in the course of sampling. Doing so compromises the validity of the sample. For more information, see Conditional sampling. |
Record Fields |
Specifies whether the output table includes the entire record, or selected fields. If you choose Fields, do one of the following: Select the field(s) to extract from the Extract Fields list. Click Extract Fields to select the field(s), or to create an expression. The order in which you select the fields is the order in which the columns appear in the results. If you are appending results to an existing Analytics table, the column selection and order must be identical to the column selection and order in the existing table. |
Subsample Report Selection Order No Repeats |
(Optional) Additional sampling options. You can select one or more of the following options:
Note Subsample is available only if Fields output is selected. Report Selection Order is available only if both the Random selection method and Fields output are selected. With No Repeats, selected records become ineligible for subsequent selection, which can reduce the size of the sample. You should consider oversampling the data set to compensate. For more information, see Sample selection without repeats. |
Append To Existing File | Specifies that the output results are appended (added) to the end of
an existing Analytics table. Note Leaving Append To Existing File deselected is recommended if you are uncertain whether the output results and the existing table have an identical data structure. For more information about appending and data structure, see Appending output results to an existing table. |
OK | Executes the operation. If the overwrite prompt appears, select the appropriate option. If you are expecting the Append option to appear and it does not, click No to cancel the operation and see Appending output results to an existing table. |
Monetary unit sampling options
For monetary unit sampling, you can specify one or more of the following options:
- Top stratum cutoff
- Subsampling
- Sample selection without repeats
Top stratum cutoff
Note
Top stratum cutoff is available only for monetary unit sampling that uses the fixed interval or the cell selection methods.
Top stratum cutoff is an additional method that Analytics uses to bias monetary unit sampling toward larger amounts. By default, sample field amounts greater than or equal to the interval value are considered top stratum amounts and are automatically included in the sample.
Negative as well as positive amounts are eligible for automatic inclusion because it is the absolute value of the amount that is considered.
Note that the greater the amount of automatic selection, the larger the sample size becomes.
You can optionally specify a top stratum cutoff value that is higher or lower than the interval value.
Top stratum cutoff value | Effect |
---|---|
Higher than the interval value |
Decreases the probability that larger amounts will be automatically included in the sample. If you specify a cutoff value larger than the largest positive or negative amount in the sample field, no amounts are automatically selected. |
Lower than the interval value |
Increases the probability that larger amounts will be automatically included in the sample. If no amounts are automatically selected using the default top stratum cutoff, you can adjust the cutoff value downward in order to automatically select some of the larger amounts in the sample field. Caution If you specify a cutoff value that is too small in relation to the sample field amounts, too many amounts are automatically selected, which defeats the purpose of sampling. |
Top stratum selections and amount recorded in the log
When you perform a monetary unit sample, the number of top stratum selections, and the total top stratum amount, is displayed in the log.
Example
The log displays that 8 of 93 selected records are top stratum, accounting for $33,153.55 of the absolute value of the numeric sample field.
Sample size = 93 (8 top stratum), out of 772 records sampled
Population: 585674.41, Top stratum: 33153.55, Other: 552520.86
Subsampling
Note
Subsampling is available only for monetary unit sampling using field output.
In some cases, each amount in a sample field represents a total of several separate transactions. If you want to perform audit procedures on only one transaction from each sampled total amount, you can use subsampling to randomly select the individual transactions.
When you select Subsample in the Sample dialog box, the resulting sample includes the SUBSAMPLE field. This field contains amounts that represent the difference between the total amount and the actual monetary unit used to select the total amount.
Example
$12,455 | (total amount) |
- $4,620 | (selected monetary unit) |
= $7,835 | (amount displayed in the SUBSAMPLE field) |
To complete the process, you would select the transaction containing the 7,835th dollar in the cumulative balance of transactions for that particular total amount.
Note
Any top stratum cutoff amounts in the sample display “0.00” in the SUBSAMPLE field because they are automatically included in the sample and no monetary unit was involved in their selection.
Sample selection without repeats
Monetary unit sampling may select the same record more than once. Each amount in the sample field contains multiple monetary units, and two or more monetary units belonging to the same amount can be selected, which means the record containing the amount is multiply selected.
You can prevent multiple selection of the same record by selecting No Repeats in the Sample dialog box. The resulting sample will not contain duplicates. However, the number of sampled records may be smaller than the sample size calculated by Analytics. To compensate, you can oversample by using one of the following methods to increase the sample size:
-
Fixed interval or cell selection methods:
- decrease the size of the interval
- adjust the top stratum cutoff value to automatically select a greater number of records
- Random selection method increase the specified sample size