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 1903.10063math.STstat.TH
收藏

增长维离散选择模型的最优线性鉴别器

Optimal Linear Discriminators For The Discrete Choice Model In Growing Dimensions

Debarghya Mukherjee, Moulinath Banerjee, Ya'acov Ritov

摘要:

离散选择模型的Manski最大得分估计是一种最优线性估计,近年来一直是计量经济学和统计学研究的热点,但其在增长维数情景下的行为还不清楚。本文件解决了这一差距。考虑两种不同的情况:$p$与$n$一起生长,但速度较慢,即$p/n \rightarrow 0$ ;和$p \gg n$(快速生长)。在二进制响应模型中,我们重铸曼斯基的得分估计为经验风险最小化的分类问题,并推导出$\ell_2$的收敛速率的得分估计下\emph{transition condition}在我们的保证金参数,校准的估计问题的难度水平。我们还建立了二元选择模型中的极小极大$\ell_2$误差的上界和下界,它们之间相差一个对数因子,并在缓慢增长的情况下构造了一个极小极大最优估计。一些扩展到一般情况下-多项式响应模型-也被认为是。最后但并非最不重要的是,我们使用各种学习算法来计算增长维度中的最大得分估计。

英文摘要:

Manski's celebrated maximum score estimator for the discrete choice model, which is an optimal linear discriminator, has been the focus of much investigation in both the econometrics and statistics literatures, but its behavior under growing dimension scenarios largely remains unknown. This paper addresses that gap. Two different cases are considered: $p$ grows with $n$ but at a slow rate, i.e. $p/n \rightarrow 0$; and $p \gg n$ (fast growth). In the binary response model, we recast Manski's score estimation as empirical risk minimization for a classification problem, and derive the $\ell_2$ rate of convergence of the score estimator under a \emph{transition condition} in terms of our margin parameter that calibrates the level of difficulty of the estimation problem. We also establish upper and lower bounds for the minimax $\ell_2$ error in the binary choice model that differ by a logarithmic factor, and construct a minimax-optimal estimator in the slow growth regime. Some extensions to the general case -- the multinomial response model -- are also considered. Last but not least, we use a variety of learning algorithms to compute the maximum score estimator in growing dimensions.

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。

↑