Statistical Sampling
Sampling can help you reach a statistically valid conclusion about a data population from a relatively small number of samples.
Analyzer supports two common sampling techniques:
|
•
|
Monetary unit sampling (MUS), in which the population consists of the absolute value of a numeric field |
|
•
|
Transaction sampling, also called record sampling, in which the population consists of the number of records |
You can choose one of these methods to select the items in the sample:
|
•
|
Fixed interval sampling |
|
•
|
Random interval sampling, also called cell sampling |
You can sample the entire population, a subset of the population or use filters to perform conditional sampling.
Analyzer has three commands for statistical sampling:
|
•
|
Size to determine statistically appropriate sample sizes and intervals |
|
•
|
Sample to draw samples from a population |
|
•
|
Evaluate to determine the effect of errors detected in your samples |
For more information on the Sample command Sample. For more information on statistical sampling see Sampling.