An empirical comparison of two novel transformation models.
An empirical comparison of two novel transformation models.
复制标题
DOI：
10.1002/sim.8425
复制
发表时间：
2020-02-28
影响因子：
2
通讯作者：
Shepherd BE
中科院分区：
文献类型：
--
作者：
Tian Y;Hothorn T;Li C;Harrell FE Jr;Shepherd BE
连续响应变量经常被转换以满足建模假设，但转换的选择可能具有挑战性。最近提出了两种转换模型：半参数累积概率模型（CPM）和参数最可能转换模型（MLT）。这两种方法都对累积分布函数进行建模，并需要指定一个链接函数，该函数隐含地假设响应在一些单调变换后遵循已知的分布。然而，这两种方法对转换的估计不同。使用CPM，拟合有序回归模型，该模型基本上将每个连续响应视为唯一类别，因此非参数估计转换; CPM是半参数线性转换模型。相比之下，对于MLT，使用灵活的基函数来参数化变换。条件期望和分位数很容易从响应变量的原始尺度上的两种方法中导出。我们比较了这两种方法与广泛的模拟。我们发现，这两种方法一般都有良好的性能与中等和大样本量。在正确的模型下，MLT在小样本量下的表现略优于CPM。CPM往往对模型错误指定和结果舍入更加稳健。除了在最简单的情况下，这两种方法都优于实践中常用的基本转换方法。我们将这两种方法应用于HIV生物标志物研究。
Continuous response variables are often transformed to meet modeling assumptions, but the choice of the transformation can be challenging. Two transformation models have recently been proposed: semiparametric cumulative probability models (CPMs) and parametric most likely transformation models (MLTs). Both approaches model the cumulative distribution function and require specifying a link function, which implicitly assumes the responses follow a known distribution after some monotonic transformation. However, the two approaches estimate the transformation differently. With CPMs, an ordinal regression model is fit, which essentially treats each continuous response as a unique category and therefore nonparametrically estimates the transformation; CPMs are semiparametric linear transformation models. In contrast, with MLTs, the transformation is parameterized using flexible basis functions. Conditional expectations and quantiles are readily derived from both methods on the response variable’s original scale. We compare the two methods with extensive simulations. We find that both methods generally have good performance with moderate and large sample sizes. MLTs slightly outperformed CPMs in small sample sizes under correct models. CPMs tended to be somewhat more robust to model misspecification and outcome rounding. Except in the simplest situations, both methods outperform basic transformation approaches commonly used in practice. We apply both methods to an HIV biomarker study.
登录
查看更多内容
DOI：
10.1111/j.2517-6161.1968.tb00724.x
发表时间：
1968-01-01
影响因子：
5.8
作者：
COX, DR;SNELL, EJ
通讯作者：
SNELL, EJ
影响因子：
2
作者：
De Neve, Jan;Thas, Olivier;Gerds, Thomas A.
通讯作者：
Gerds, Thomas A.
影响因子：
1.2
作者：
Manuguerra, Maurizio;Heller, Gillian Z.
通讯作者：
Heller, Gillian Z.
DOI：
10.1002/cjs.11302
发表时间：
2016-12
期刊：
The Canadian journal of statistics = Revue canadienne de statistique
影响因子：
--
作者：
Shepherd BE;Li C;Liu Q
通讯作者：
Liu Q
影响因子：
1
作者：
Hothorn, Torsten;Moest, Lisa;Buehlmann, Peter
通讯作者：
Buehlmann, Peter