1Section of Economics and Statistics, Indian Institute of Horticultural Research, Hessaraghattn Lake Post Office, Bangalore-560089, Karnataka.
2Division of Floriculture and Landscaping, Indian Agricultural Research Institute, New Delhi-110012
Statistical models are widely used in the area of horticulture with a view to express the functional relationship among a dependent variable and a gamete of explanatory variables. It is essential to make use of sound statistical principle and work out the required goodness-of-statistics while taking decision about the adequacy of the identical model. An attempt has been made in this article to fit a suitable statistical model in order to determine the optimum number of leaves to be retained on plant while harvesting of cut spike at commercial stage in gladiolus cv. Pink friendship. In particular, multiple regression models of first and second degrees were developed separately for 13 different quality characteristics of yield and quality of corm and 21 different vegetative growth flowering and corm development characteristics in forthcoming crop. Based on the estimated and tested values of co-efficient of determination computed for different model fits, it was concluded that by and large five to six leaves were found to be optimum in estimating the characters under study.
Gladiolus, number of leaves, goodness-of-fit statistics, regression model
