Testing for Equivalence: A Methodology for Computational Cognitive Modelling
Testing for Equivalence: A Methodology for Computational Cognitive Modelling
复制标题
等价性测试：计算认知建模的方法
DOI：
--
复制
发表时间：
2010
期刊：
影响因子：
--
通讯作者：
R. West
中科院分区：
文献类型：
--
作者：
T. Stewart;R. West
等效性测试（Stewart and West, 2007; Stewart, 2007）是一种评估模型与被建模系统之间相似性的统计度量。它的设计是为了避免过度拟合，并生成一个易于解释的模型质量摘要。我们将等价检验应用于两个任务：重复二元选择（Erev et al., 2010）和动态库存和流量（Gonzalez and Dutt, 2007）。在第一种情况下，我们找到了广泛的统计等效模型（并赢得了预测竞赛），同时确定了尚未充分捕获的任务的特定方面。在第二种情况下，我们重新评估动态库存和流量挑战的结果，展示我们的方法如何强调模型覆盖的广度，以及如何使用它来比较不同的模型。我们认为，模型的解释力取决于数值相似性的经验数据在一套广泛的措施。
Testing for Equivalence: A Methodology for Computational Cognitive Modelling The equivalence test (Stewart and West, 2007; Stewart, 2007) is a statistical measure for evaluating the similarity between a model and the system being modelled. It is designed to avoid over-fitting and to generate an easily interpretable summary of the quality of a model. We apply the equivalence test to two tasks: Repeated Binary Choice (Erev et al., 2010) and Dynamic Stocks and Flows (Gonzalez and Dutt, 2007). In the first case, we find a broad range of statistically equivalent models (and win a prediction competition) while identifying particular aspects of the task that are not yet adequately captured. In the second case, we re-evaluate results from the Dynamic Stocks and Flows challenge, demonstrating how our method emphasizes the breadth of coverage of a model and how it can be used for comparing different models. We argue that the explanatory power of models hinges on numerical similarity to empirical data over a broad set of measures.