arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1901.04376stat.ME
收藏

使用准经验残差分布函数评估具有离散结果的回归模型

Assessment of Regression Models with Discrete Outcomes Using Quasi-Empirical Residual Distribution Functions

Lu Yang

摘要:

对于离散结果的回归模型,对模型的充分性做出明智的决策一直是一个突出的问题。这些结果的标准评估工具(例如偏差残差)通常显示出与假设模式的较大差异,即使在真实模型下也是如此,并且不提供信息,特别是当数据高度离散时(例如二进制)。为了填补这一空白,我们提出了一个准经验残差分布函数一般离散(如序数和计数)的结果,作为替代的经验Cox-Snell残差分布函数。我们提出的评估工具是一种原则性的方法,不需要向数据中注入噪声。当至少有一个连续的协变量是可用的,我们渐近地表明,所提出的函数一致收敛到正确指定的模型下的身份函数,即使具有高度离散的结果。通过模拟研究,我们实证表明,拟议的准经验残差分布函数优于常用的各种模型评估任务的残差,因为它是接近真实模型下的假设模式,并显着偏离此模式下的模型误指定,因此是一个有效的评估工具。

英文摘要:

Making informed decisions about model adequacy has been an outstanding issue for regression models with discrete outcomes. Standard assessment tools for such outcomes (e.g. deviance residuals) often show a large discrepancy from the hypothesized pattern even under the true model and are not informative especially when data are highly discrete (e.g. binary). To fill this gap, we propose a quasi-empirical residual distribution function for general discrete (e.g. ordinal and count) outcomes that serves as an alternative to the empirical Cox-Snell residual distribution function. The assessment tool we propose is a principled approach and does not require injecting noise into the data. When at least one continuous covariate is available, we show asymptotically that the proposed function converges uniformly to the identity function under the correctly specified model, even with highly discrete outcomes. Through simulation studies, we demonstrate empirically that the proposed quasi-empirical residual distribution function outperforms commonly used residuals for various model assessment tasks, since it is close to the hypothesized pattern under the true model and significantly departs from this pattern under model misspecification, and is thus an effective assessment tool.

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。

↑