Name
Title | ||
|---|---|---|
Theoretical Analysis and Comparison of Several Criteria on Linear Model Dimension Reduction |
Abstract | ||
|---|---|---|
Detecting the dimension of the latent subspace of a linear model, such as Factor Analysis, is a well-known model selection problem. The common approach is a two-phase implementation with the help of an information criterion. Aiming at a theoretical analysis and comparison of different criteria, we formulate a tool to obtain an order of their approximate underestimation-tendencies, i.e., AIC, BIC/MDL, CAIC, BYY-FA(a), from weak to strong under mild conditions, by studying a key statistic and a crucial but unknown indicator set. We also find that DNLL favors cases with slightly dispersed signal and noise eigenvalues. Simulations agree with the theoretical results, and also indicate the advantage of BYY-FA(b) in the cases of small sample size and large noise. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1007/978-3-642-00599-2_20 | ICA |
Keywords | Field | DocType |
approximate underestimation-tendencies,theoretical analysis,noise eigenvalues,linear model dimension reduction,large noise,theoretical result,well-known model selection problem,different criterion,linear model,common approach,factor analysis,dimension reduction,model selection,eigenvalues | Applied mathematics,Mathematical optimization,Dimensionality reduction,Statistic,Subspace topology,Linear model,Minimum description length,Model selection,Sample size determination,Eigenvalues and eigenvectors,Mathematics | Conference |
Volume | ISSN | Citations |
5441 | 0302-9743 | 2 |
PageRank | References | Authors |
0.44 | 8 | 2 |