Informatics Publishing Limited

Ajay Verma ¹, V. Kumar ¹, A. S. Kharab ¹, G. P. Singh ¹

Affiliations
1 Statistics and Computer Center, ICAR-Indian Institute of Wheat and Barley Research, Karnal (Haryana), IN

Abstract

The present study was carried out to identify malt barley genotypes with high yield and stability across eight different environments, using non-parametric statistical measures. Descriptive statistics MR, SD and CV identified DWRB147, DWRB150 and RD2943 stable genotypes. BH902 and PL890 were identified as unstable genotypes by CMR CSD and CCV. Non-parametric measures selected DWRB147 and DWRB150 as the stable genotypes and BH902 and PL890 unstable genotypes. Significant tests for S_i ¹ and S_i ² were based on sum of Z_i ¹ and Z_i ² measures and sum of Z_i ¹ was greater than critical value confirmed significant differences among the twenty genotypes. Results of the NP_i ², NP_i ³ and NP_i ⁴were similar for unstable performance of BH902, DWRB150 and DWRB147. Biplot analysis of PCA1 and PCA2 accounting for 70.08 per cent showed three distinguish groups among non-parametric measures. Clustering by Ward’s hierarchical method expressed four clusters by using the squared Euclidean distance as dissimilarity measure.

Keywords

Non-Parametric Measurements, Rank Correlation, Biplot Analysis, Hierarchical Clustering.

Full Text

     

Ajay Verma ¹, V. Kumar ¹, A. S. Kharab ¹, G. P. Singh ¹

Affiliations
1 Statistics and Computer Center, ICAR- Indian Institute of Wheat and Barley Research, Karnal (Haryana), IN

Abstract

Genotypes G5, G8, G3, G21 and G18 had achieved higher yields besides b_i > 1.0. G21 and G3 identified as appropriate one, because had higher yield value than the mean, bi values near 1.0 and low S²_di. Lower values (W²_i) resulted for G12, G5, G2, G21 while higher for G5, G3 and G14. Genotypes G12 followed by G2, G20, and G7 had the smallest environmental variance (S²_xi). Smaller values of (CV_i) considered G12, G2, G20, and G10 of stable performance. α² _i measure pointed out G12, G7 and G2 with smallest values. Desirable lower P_i values reflected by G18, G5, G21, and G4 while GAI values identified G18, G11, G4 G10 as desirable genotypes. S_i ⁽¹⁾ and S_i⁽²⁾ showed lower values of G12, G2 and G7 genotypes. Significant tests of S_i ⁽¹⁾ and S_i⁽²⁾ proved the highly significant difference in ranks among the 21 genotypes grown in 8 environments. Genotypes G12, G2, and G7 had the lower S_i⁽³⁾ and S_i⁽⁶⁾ values. Yield of genotypes had significant negative correlation with bi, S_i⁽²⁾, S_i⁽³⁾, S_i⁽⁶⁾, NP_i ⁽²⁾, NP_i⁽³⁾, NP_i⁽⁴⁾ and significant positive correlation with GAI, P_i and Rank Sum. Hierarchical cluster analysis classified genotypes into three clusters as largest cluster included genotypes with more than average yield along with high yielders G18, G11, G3, G5, G21 and unstable performance indicated by non parametric measures. Biplot analysis while considering first two significant principal components grouped the parametric and non parametric measures into four groups. The smaller group consisted of b_i and S² _di and adjacent to group of non parametric measures S_i⁽²⁾, S_i⁽⁶⁾, NP_i⁽²⁾, NP_i⁽³⁾ and NP_i⁽⁴⁾.

Keywords

Barley, Parametric, Non-Parametric Measures, Biplot Analysis, Hierarchical Clustering.

Full Text