Cases for the nugget in modeling computer experiments
Cases for the nugget in modeling computer experiments
复制标题
DOI：
10.1007/s11222-010-9224-x
复制
发表时间：
2012-05-01
影响因子：
2.2
通讯作者：
Lee, Herbert K. H.
中科院分区：
文献类型：
--
作者：
Gramacy, Robert B.;Lee, Herbert K. H.
大多数计算机实验的代理模型都是内插器，而最常见的内插器是高斯过程(GP)，它故意省略了一个称为块块的小尺度(测量)误差项。其解释是，根据定义，计算机实验是“确定性的”，因此不存在测量误差。我们认为，对于计算机实验来说，这是一个过于狭隘的焦点，也是一种在统计学上低效的建模方式。我们表明，在各种常见情况下，估计(非零)块可以导致具有更好的统计特性的代理模型，例如预测精度和覆盖率。
Most surrogate models for computer experiments are interpolators, and the most common interpolator is a Gaussian process (GP) that deliberately omits a small-scale (measurement) error term called the nugget. The explanation is that computer experiments are, by definition, "deterministic", and so there is no measurement error. We think this is too narrow a focus for a computer experiment and a statistically inefficient way to model them. We show that estimating a (non-zero) nugget can lead to surrogate models with better statistical properties, such as predictive accuracy and coverage, in a variety of common situations.