ISSN : 2320-7051

  • No. 772, Basant Vihar, Kota

    Rajasthan-324009 India

  • Call Us On

    +91 9784677044


International Journal of Pure & Applied Bioscience (IJPAB)
Year : 2018, Volume : 6, Issue : 6
First page : (248) Last page : (258)
Article doi: :

Use of Ordinal Logistic Regression and Multiclass Discriminant Model for Classification of Genotypes for Maturity of Little Millet

Nagaraja M. S.1 and Abhishek Singh2

1Assistant Professor, Department of Statistics, Christ (Deemed To Be University),
Hosur Road, Bengaluru-560029
2Assistant Professor, Department of Farm Engineering, Institute of Agricultural Sciences,
Banaras Hindu University, Uttar Pradesh-221005
*Corresponding Author E-mail:
Received: 14.02.2018 | Revised: 23.03.2018 | Accepted: 3.04.2018  



Classification in agricultural systems are quite useful for Planning purposes for which various subjective and objective approaches are in vogue, classification of genotypes or germplasm based on yield and yield attributing characters is important for an accurate measurement of the differences between populations as well as for rapid assessment of their breeding potential. In present study the effort has been made to study the statistical model such as Ordinal logistic regression model and Multiclass Discriminant model and same has been used for classification of genotypes of little millet for different classes of maturity based on yield and yield attributing characters. These models were fitted to secondary data recorded on yield and yield attributing characters of 722 genotypes of little millet and the data has been collected from Project coordination cell, All India Coordinated Small Millets Improvement Project (AICSMIP), ICAR, and Bengaluru. Classes of Fifty percent flowering (Maturity) was considered as dependent variable and all other attributing characters as predictors. Classification ability measures such as Accuracy Rate, Kappa Statistics, Avgprecision, and Avgrecall were used for testing samples. Yield, Plant height, Number of basel tillers, Flag leaf length, Flag leaf width were considered to be important attributing characters for classification and Multiclass Discriminant model (71.72 %) was performed compare to Ordinal Logistic Regression Model (68.28 %) for both classification of genotypes for different classes of maturity of little millets.

Key words: Ordinal logistic regression, Multiclass Discriminant model, Classification, Attributing, Accuracy.

Full Text : PDF; Journal doi :

Cite this article: Nagaraja, M.S. and Singh, A., Use of Ordinal Logistic Regression and Multiclass Discriminant model for Classification of genotypes for maturity of Little Millet, Int. J. Pure App. Biosci.6(6): 248-258 (2018). doi: