欢迎访问《中国农学通报》,

中国农学通报 ›› 2016, Vol. 32 ›› Issue (30): 14-21.doi: 10.11924/j.issn.1000-6850.casb16060113

所属专题: 小麦

• 农学 农业基础科学 • 上一篇    下一篇

基于R语言的冬小麦品种稳定性分析

盛 坤1,2,杨丽娟1,李晓航1,王映红1,蒋志凯1   

  1. (1河南省新乡市农业科学院,河南新乡 453000;2中国农业科学院农田灌溉研究所,河南新乡 453000)
  • 收稿日期:2016-06-22 修回日期:2016-10-15 接受日期:2016-08-25 出版日期:2016-10-31 发布日期:2016-10-31
  • 通讯作者: 盛 坤
  • 基金资助:
    国家现代农业产业技术体系建设专项资金项目“小麦隐性灾害防控”(CARS-3-2-35);河南省重大科技专项“小麦新品种选育与示范”(151100110400)。

Stability Analysis of Winter Wheat Varieties Based on R Language

Sheng Kun1,2, Yang Lijuan1, Li Xiaohang1, Wang Yinghong1, Jiang Zhikai1   

  1. (1XinXiang Academy of Agricultural Sciences, Xinxiang Henan 453000; 2Farmland Irrigation Research Institute, Chinese Academy of Agricultural Sciences, Xinxiang Henan 453000)
  • Received:2016-06-22 Revised:2016-10-15 Accepted:2016-08-25 Online:2016-10-31 Published:2016-10-31

摘要: 旨在确定适宜的品种稳定性分析参数及其计算方法。利用冬小麦区域试验数据,通过R语言计算了14个稳定性参数,并采用秩相关和主成分分析,研究了参数的功效和相关性。结果表明,R语言可以方便、灵活的计算各类参数;相关分析表明,PCOA和产量极显著正相关,CVBi极显著正相关,Sigma.squareSi、Si2、Si3、ASV极显著正相关,ASVSi6极显著正相关;主成分分析表明,14个稳定性参数可以分为3组,组1包含Bi,与多数参数负相关,与产量相关性不显著,组2包含YIELD、Pi、YSI、CVPCOA,与产量显著正相关,组3包含Sigma.square、S.square、ASV、Si、Si2、Si3、Wi、Si6,与产量相关性不显著。研究表明,ASVSi与其他参数间的相关性较好,PCOA模型能够同时评价品种产量及其稳定性,三者是单变量参数模型的有益补充。

关键词: 薇甘菊, 薇甘菊, 调查监测, 风险分析

Abstract: This article aimed at determining the suitable variety stability analysis parameters and their calculation methods. Fourteen parameters for one set of variety regional test data were computed by R language. The effect and correlation of the parameters were studied using rank correlation and principal component analysis. The results showed that R language was a convenient and flexible language for calculating all kinds of parameters. Correlation analysis showed that the correlations between PCOA and yield, CV and Bi, Sigma.square and Si, Si2, Si3, ASV, respectively, ASV and Si6 were positive significantly. The principal components analysis showed that 14 methods could be categorized in three groups: group Ⅰ included Bi which showed negative correlation with others and no correlation with yield; group Ⅱ YIELD, Pi, YSI, CV and PCOA whose yield level and stability of performance were considered simultaneously; and group Ⅲ which were mostly associated with each other and showed little or no correlation with yield. The research showed that the ASV, Si and the PCOA were the beneficial supplements of the single variable parameter model.