In multivariate regression, when covariates are numerous, it is often reasonable to assume that only a small number of them has predictive information. In some medical applications for instance, it is believed that only a few genes out of thousands are responsible for cancers. In that case, the aim is not only to propose a good fit, but also to select the relevant covariates (genes). We propose to perform model selection with additive models in high dimensions (sample size and number of covariates). Our approach is computationally efficient thanks to fast wavelet transforms, it does not rely on cross validation, and it solves a convex optimization problem for a prescribed penalty parameter, called the quantile universal threshold. We also propose a second rule based on Stein unbiased risk estimation geared towards prediction. We use Monte Carlo simulations and real data to compare various methods based on false discovery rate (FDR), true positive rate (TPR) and mean squared error. Our approach is the only one to handle high dimensions, and has the best FDR--TPR trade-off.
翻译:在多变量回归中,当共变数众多时,通常可以合理地假定只有一小部分人具备预测信息。例如,在某些医疗应用中,人们认为只有数千个基因中的少数基因才对癌症负责。在这种情况下,目的不仅是提出一个良好的适配,而且还要选择相关的共变数(基因)。我们提议采用高维(抽样大小和共变数)的添加模型进行模型选择。我们的方法是通过快速波盘变换来计算效率高,它不依赖交叉验证,它解决了指定惩罚参数的螺旋优化问题,称为通用临界值。我们还提出了以预测为目的的基于不偏差风险估算的第二个规则。我们使用蒙特卡洛模拟和真实数据来比较基于虚假发现率、真实正率和平均平方差的各种方法。我们的方法只是处理高维值,并且拥有最佳的FDR-TR交易。