arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 0709.3526math.STstat.TH
收藏

log-线性群-Lasso估计及其渐近性质

The log-linear group-lasso estimator and its asymptotic properties

Yuval Nardi, Alessandro Rinaldo

摘要:

在多项抽样方案下,我们定义了表示多层对数线性模型的指数分布族的自然参数的群Lasso估计。这样的估计出现作为基于组套索罚的凸惩罚似然最优化问题的解决方案。我们说明了如何有可能构建一个估计的基础日志线性模型使用块的非零系数恢复组套索过程。我们调查的渐近性质的组套索估计作为一个模型选择方法在一个双渐近的框架,其中的样本量和模型的复杂性同时增长。我们提供的条件,保证组套索估计是模型选择一致的,在这个意义上说,压倒性的概率作为样本量的增加,它正确地识别所有的非零变量之间的相互作用。如果真实基础模型的序列足够稀疏,即使细胞数量增长大于样本量,恢复也是可能的。最后,我们得到了log-线性群-lasso估计的中心极限型结果.

英文摘要:

We define the group-lasso estimator for the natural parameters of the exponential families of distributions representing hierarchical log-linear models under multinomial sampling scheme. Such estimator arises as the solution of a convex penalized likelihood optimization problem based on the group-lasso penalty. We illustrate how it is possible to construct an estimator of the underlying log-linear model using the blocks of nonzero coefficients recovered by the group-lasso procedure. We investigate the asymptotic properties of the group-lasso estimator as a model selection method in a double-asymptotic framework, in which both the sample size and the model complexity grow simultaneously. We provide conditions guaranteeing that the group-lasso estimator is model selection consistent, in the sense that, with overwhelming probability as the sample size increases, it correctly identifies all the sets of nonzero interactions among the variables. Provided the sequences of true underlying models is sparse enough, recovery is possible even if the number of cells grows larger than the sample size. Finally, we derive some central limit type of results for the log-linear group-lasso estimator.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑