Comparative study on different nonparametric stability measures in soybean

Abstract

The assessment of stable genotypes to a wide range of environments is important for recommending cultivars in plant breeding programme. Therefore, twenty four genotypes of soybean were tested over three environments in Tamil Nadu. The stability analysis of GxE interaction observed data is mostly handled by parametric methods. If any assumptions of parametric methods are violated, the result of these methods may be questionable. The nonparametric measures are easy to analyse and simple to interpret which have more advantages than parametric methods. In this paper, nine nonparametric methods were used for identification of stable genotype and association among these measures were checked by correlation. According to Principle component analysis, nonparametric measures were divided into three groups Group 1 included Kang ranksum, NP_i⁽¹⁾, NP_i⁽²⁾, NP_i⁽³⁾ and NP_i⁽⁴⁾, Group 2 included S_i⁽³⁾ and S_i⁽⁶⁾. Group 3 included S_i⁽¹⁾ and S_i⁽²⁾. According to nonparametric measures, G17 is the stable one.