Audit software for extracting samples (no cost). View description of algorithm used. Example in Excel file.

Only debits are subject to sampling. The random start can either be specified, or else can be computed by the system (if not specified). A reliability factor (“R”) of either 1,2 or 3 must be provided. The sampling interval (“J”) is also provided. The system will then divide the “J” amount by the R value to determine the sampling interval to be performed. The procedure writes a reconciliation report at the end of processed to tie the samples selected to the population totals.

Although credits are not sampled, the total number and amount of credit transactions is reported.

Cumulative Monetary Amount (CMA) ScreenShot

Cumulative Monetary Amount (CMA) Reconciliation Report

A reconciliation report is output automatically. Below is an example.

Syntax - CMA Sample
proc sample data=DATAFILE type=cma r=R j=J rn=AMOUNT;
out = OUTPUTRESULT;
var VARNAME;

Parameters Used - CMA Sample
There are six parameters for the CMA sample procedure:

DATAFILE - the datafile to be analyzed
SAMPLETYPE - always "cma"
R - R Factor which is 1,2 or 3 (1 =99%, 2=95%, 3=67%)
J - J Numeric Amount to be used as interval
AMOUNT – the random starting amount,
VARNAME - the numeric field, which is the amount to be tested

Parameter Values
The R factor must be one of either 1,2 or 3. If not, the procedure will terminate in error.
The sampling interval is computed as J / R. For example, if J is 12,000,000 and R is 3, then the sampling interval will be 4,000,000.
The Random starting amount can either be specified as zero or an amount between zero and J / R. If omitted, the value will be zero. If the amount is specified as zero, then the system will compute a random starting amount. If it is specified as a non-zero amount, then that amount will be used if it lies between zero and (J / R). Note that the amount is specified as a positive number, but the random starting amount will be a negative value.

Example Script - CMA Sample

*;
* csample5.ezs;
* CMA sample procedure;
* select a cma sample and prepare the reconciliation schedule;
libname test '{%libout}';
libname fs '{%libin}';
proc sample data=fs.drcrpop type=cma r=1 j=50000 rn=23742;
out = test.csamp5;
var tran;
run;

The example script above, specified that the population is to be sampled as follows:
The population is the records contained in the directory specified by %libin, with the name of drcrpop.tab. An “R” factor of 1 has been specified, i.e. 99% reliability level.
A “J” factor of $50,000 has been specified.
A random start has been specified as 23742. The sampling interval is 50000 = (50000 / 1) so the randm start specified is valid.
The sample extract is to be written to the library specified by %libout, with a name of “csamp5.tab”. In addition, a reconciliation report will be written in the same directory with a name of csamp5.rec.

View an overview of EZ-R Stats for Windows. Download EZ-R Stats for Windows and other products.

Analysts interested in Proc Sample may also be interested in Proc Means and Proc Univariate.

There is also a web analytics version which can be run directly from the Internet using Excel or other data source (in tab separated value format). View Web Analytics details. All software is provided at no cost.