Name
Abstract | ||
|---|---|---|
Penalized regression models are popularly used in high-dimensional data analysis to conduct variable selection and model fitting simultaneously. Whereas success has been widely reported in literature, their performances largely depend on the tuning parameters that balance the trade-off between model fitting and model sparsity. Existing tuning criteria mainly follow the route of minimizing the estimated prediction error or maximizing the posterior model probability, such as cross validation, AIC and BIC. This article introduces a general tuning parameter selection criterion based on variable selection stability. The key idea is to select the tuning parameters so that the resultant penalized regression model is stable in variable selection. The asymptotic selection consistency is established for both fixed and diverging dimensions. Its effectiveness is also demonstrated in a variety of simulated examples as well as an application to the prostate cancer data. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.5555/2567709.2567772 | Journal of Machine Learning Research |
Keywords | Field | DocType |
model fitting,tuning parameter,asymptotic selection consistency,penalized regression model,model sparsity,variable selection stability,posterior model probability,resultant penalized regression model,consistent selection,general tuning parameter selection,variable selection,tuning,kappa coefficient,stability | Mean squared prediction error,Pattern recognition,Feature selection,Regression analysis,Selection criterion,Cohen's kappa,Artificial intelligence,Cross-validation,Mathematics | Journal |
Volume | Issue | ISSN |
14 | Issue-in-Progress | Journal of Machine Learning Research 2013, Vol. 14, 3419-3440 |
Citations | PageRank | References |
11 | 0.78 | 5 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| SUN Wei | 1 | 247 | 26.63 |
| Junhui Wang | 2 | 267 | 13.46 |
| Yixin Fang | 3 | 40 | 4.79 |