Model-Dependent Variance Inflation Factor Cutoff Values

Article

Craney, Trevor A.; Surles, James G.   (2002, ASQ and Marcel Dekker, Inc.)   Pratt & Whitney, East Hartford, CT; Texas Tech University, Lubbock, TX (craneyt@pweh.com)

Quality Engineering    Vol. 14    No. 3
QICID: 18072    March 2002    pp. 391-403
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

[This abstract is based on the authors' abstract.] In running designed experiments, it can be difficult to retain the precise settings required to maintain orthogonal effects. This is not considered to be a measurement error if the settings can be fine-tuned once the experiment is under way. A comparison is made on a design using both the intended settings and actual settings. Induced collinearity in the effects is measured using variance inflation factors. Two cutoff values are proposed to determine when an effect's variance is too great to keep that effect in the model.

Keywords

Case study,Collinearity,Precision,Measurement error,Orthogonal array (OA),Design of experiments (DOE)


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.