Distribution Analysis: Location and Scale Statistics


Overview of the Location and Scale Statistics Analysis

Univariate data are often summarized by computing statistics that estimate location and scale. The mean, median, mode, trimmed mean, and Winsorized mean are all statistics that describe the location (or central tendency) of data. Statistics that describe the scale (or variability) include the standard deviation, interquartile range, Gini’s mean difference, and median absolute deviation from the median (MAD). You can use the Location and Scale Statistics analysis to compute location and scale estimates for a single numeric variable. You can also test the hypothesis that the population mean equals a particular value.

You can run a Location and Scale Statistics analysis by selecting AnalysisDistribution AnalysisLocation and Scale Statistics from the main menu. When you request location and scale estimates, SAS/IML Studio calls the UNIVARIATE procedure in Base SAS software. See the UNIVARIATE procedure documentation in the Base SAS Procedures Guide for additional details.