Design of variable resolution for model selection

作者：Lin C Devon^*; Morrill Susanne

来源：Journal of Statistical Planning and Inference, 2014, 155: 127-134.

DOI：10.1016/j.jspi.2014.06.001

• Bayesian variable selection
• Dantzig selector
• Generalized minimum aberration
• Hadamard matrix
• Interaction
• Orthogonal array

Designs of variable resolution were introduced for practical situations where interactions arise only within certain factors. Optimal designs of variable resolution based on the minimum G(2)-aberration criterion and design efficiency are provided. A simulation study is carried out to demonstrate the merit of designs of variable resolution in model selection. Some comments are made on the performance of optimal designs of variable resolution in using certain model selection methods.

• 出版日期2014-12