期刊
JOURNAL OF TESTING AND EVALUATION
卷 47, 期 1, 页码 449-458出版社
AMER SOC TESTING MATERIALS
DOI: 10.1520/JTE20170461
关键词
comparison; regression; sign test; test of hypothesis; Wilcoxon test
Linear and nonlinear regression models are flexible methods of data analysis that may be appropriate whenever a quantitative response variable is to be examined in relationship to any other explanatory variables. This relationship can be expressed by different models and equations. In some fields, such as in agriculture, biology, hydrology, neural network, and psychology, researchers need to analyze whether the relationship between response variable and predictor variables differ in two fitted models on the same dataset. In other words, we are interested in the comparison of two regression models for a single dataset. In this article, we will use the nonparametric methods to establish hypothesis testing for the equality of two dependent regression models. Then, a simulation study is provided to investigate the performance of the proposed method. Also, the proposed method is applied to compare the different linear, quadratic, cubic, and exponential models that can be fitted on a real dataset.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据