Model combining in factorial data analysis
We study the properties of a model combining method, ARM (adaptive regression by mixing), in the ANOVA framework. We propose model instability measures as a guide to the appropriateness of model combining in applications. We further systematically investigate the relationship between ARM performance and the underlying model structure. We propose an approach to evaluating the importance of factors based on the combined estimates. A theoretical risk bound on the combined estimator is also obtained.
This preprint was published as Lihua Chen, Panayotis Giannakouros, Yuhong Yang, "Model Combining in Factorial Data Analysis", Journal of Statistical Planning and Inference (2007): 2920-2934, doi: 10.1016/j.jspi.2006.10.005.