Column: Statistics Roundtable: Another Data Mining Tool

Article

Mason, Robert L.; Young, John C.   (2002, ASQ)   Southwest Research Institute in San Antonio, TX; InControl Technologies in Lake Charles, LA;

Quality Progress    Vol. 36    No. 2
QICID: 18841    February 2003    pp. 76-79
List $10.00
Member $5.00

FOR A LIMITED TIME, ACCESS TO THIS CONTENT IS FREE!
You will need to be signed in.
New to ASQ? Register here.

Article Abstract

In this column, we introduce the use of a Hotelling's T2 statistic as a data mining tool for large and small data sets composed of many variables. We will show how the T2 statistic, based on a single p-dimensional observation vector (x1, x2, ..., xp), is quite adept at detecting data trends, data patterns and, in general, atypical observations far removed from the majority of other observations. Subgroups of observations with a different correlation structure are easily identified, as are observations with components that contain extreme values. Further-more, we will discuss a procedure that can be used to quickly locate the source of the trend in terms of the variables under consideration.

Keywords

Hotelling's T2 statistic,Data analysis


Browse QIC Articles Chronologically:     Previous Article     Next Article

New Search

Featured advertisers





ASQ is a global community of people passionate about quality, who use the tools, their ideas and expertise to make our world work better. ASQ: The Global Voice of Quality.