Final interaction models were selected and evaluated by permutation test and/or cross-validation test. Both the multifactor-dimensionality reduction(MDR) and classification and regression trees(CART) methods revealed a high-order gene-gene interaction among KCNMB1,RGS2,PRKG,and MYLK genes (P value of permutation in MDR=0.012).
多因子降维法和分类回归树法都显示了RGS2、PRKG1、KCNMB1和MYLK基因对于高血压危险存在一种显著的高阶基因-基因交互作用(多因子降维法得到的置换检验P值=0.012)。
参考来源 - Renalase基因和血管平滑肌细胞收缩通路相关基因多态性与原发性高血压的关联研究An improved method based on permutation test was proposed,whose performance was investigated by two real examples.
之后将置换检验的方法引入信用评分模型的验证,通过两个实例的实证研究,考察了置换检验在二项响应回归模型预测能力中的稳定表现。
参考来源 - 统计方法在信用卡风险管理中的若干应用·2,447,543篇论文数据,部分数据来源于NoteExpress
应用推荐