|Model selection through a statistical analysis of the minimum of a weighted least squares cost function|de Brauwere, A.; De Ridder, F.; Pintelon, R.; Elskens, M.; Schoukens, J.; Baeyens, W.F.J. (2005). Model selection through a statistical analysis of the minimum of a weighted least squares cost function. Chemometr. Intell. Lab. Syst. 76(2): 163-173. dx.doi.org/10.1016/j.chemolab.2004.10.006
In: Chemometrics and Intelligent Laboratory Systems. Elsevier: Amsterdam; New York; Oxford; Tokyo. ISSN 0169-7439, more
|Also published as |
- de Brauwere, A.; De Ridder, F.; Pintelon, R.; Elskens, M.; Schoukens, J.; Baeyens, W.F.J. (2007). Model selection through a statistical analysis of the minimum of a weighted least squares cost function, in: VLIZ Coll. Rep. 35-36(2005-2006). VLIZ Collected Reprints: Marine and Coastal Research in Flanders, 35-36: pp. Chapter 15, more
|Authors|| || Top |
- de Brauwere, A., more
- De Ridder, F., more
- Pintelon, R.
- Elskens, M., more
- Schoukens, J.
- Baeyens, W.F.J., more
Combining (i) a statistical interpretation of the minimum of a Weighted Least Squares cost function and (ii) the principle of parsimony,a model selection strategy is proposed. First, it is compared via simulation to model selection methods based on information criteria (AICand MDL type). The first kind of simulations shows that the cost function approach outperforms in selecting the true model, especiallywhen the number of data is very small compared with the number of parameters to be estimated. Next, the model metaselection proposedby de Luna and Skouras [X. De Luna, K. Skouras, Choosing a model selection strategy, Scand. J. Stat. 30(1) (2003) 113-128.] isemployed as an objective method to choose the best model selection method. Applied to one of their examples, clearly the cost functionstrategy is selected as the best method. Finally, on a set of field data, the cost function approach is used for selecting the relevantparameters of a complex model.