Design matrix

Decision Type:

Method Category:

Categories: Terms

In design of experiments, a matrix where the rows represent individual experimental treatments. Depending on the designed experiment, it may be possible to extend the matrix using deduced levels of other functions of the factor levels (such as interaction, quadratic terms, etc.

Conducting the Experiment

𝐴-optimal design
An optimal design that maximizes the trace of 𝑋′𝑋, where 𝑋 is the design matrix. This is useful to minimize the sum of the variances of the different regression coefficient estimators.
𝒟-optimal design
An optimal design that maximizes the determinant of 𝑋′𝑋, where 𝑋 is the design matrix. Useful to produce high precision regression parameter estimates.

« Back to Glossary Index

About Fred Schenkelberg

I am the reliability expert at FMS Reliability, a reliability engineering and management consulting firm I founded in 2004. I left Hewlett Packard (HP)’s Reliability Team, where I helped create a culture of reliability across the corporation, to assist other organizations.

About Fred Schenkelberg

Leave a Reply Cancel reply