E(s2)-Optimality and Minimum Discrepancy in 2-Level Supersaturated Designs

Min Qian Liu, Fred J. Hickernell

Research output: Contribution to journalJournal articlepeer-review


Supersaturated experimental designs are often assessed by the E(s2) criterion, and some methods have been found for constructing E(s2)-optimal designs. Another criterion for assessing experimental designs is discrepancy, of which there are several different kinds. The discrepancy measures how much the empirical distribution of the design points deviates from the uniform distribution. Here it is shown that for 2-level supersaturated designs the E(s2) criterion and a certain discrepancy share the same optimal designs.
Original languageEnglish
Pages (from-to)931-939
Number of pages9
JournalStatistica Sinica
Issue number3
Publication statusPublished - Jul 2002
Externally publishedYes

User-Defined Keywords

  • Hamming distance
  • reproducing kernel
  • uniformity


Dive into the research topics of 'E(s2)-Optimality and Minimum Discrepancy in 2-Level Supersaturated Designs'. Together they form a unique fingerprint.

Cite this