Vol. 10 (2) : March-April 2019 issue
Green Farming Vol. 10 (2) : 249-252 ; March-April, 2019
Non-linear statistical modeling of area and production of banana in Coimbatore
GOMATHI T.a1, VASANTHI R.b2*, PADMARANI S.c3, S.G. PATILb4 and KALPANA M.d5
aDepartment of Agricultural Statistics, bDepartment of Physical Sciences & Information Technology, cDepartment of Agricultural Economics, Tamil Nadu Agricultural University,
Coimbatore - 641 003 (Tamil Nadu),
dDepartment of Social Science, Anbil Dharmalingam Agricultural College and Research Institute (TNAU), Navalur Kuttappattu, Tiruchirappalli - 620 027 (Tamil Nadu)
Designation : 1P.G. Student, 2,3,4,5Asstt. Professor *(vasanthi@tnau.ac.in)
Subject : Agriculture Economics, Agri-Business, Marketing & Statistics, Farm Management
Paper No. : P-7574
Total Pages : 4
Received : 28 March 2019
Revised accepted : 29 March 2019
Get Access
Citation :
GOMATHI T., VASANTHI R., PADMARANI S., S.G. PATIL and KALPANA M. 2019. Non-linear statistical modeling of area and production of banana in Coimbatore. Green Farming Vol. 10 (2) : 249-252 ; March-April, 2019
ABSTRACT
This study aim to identify the best model for area and production of banana in Coimbatore. A secondary data was collected over a period of 30 years. There are six nonlinear models such as Sinusoidal, Rational, Gaussian, Hoerl, Gompertz, and Logistic model is used. The best model is selected based on R2, RMSE and MAE values. According to area & production of banana, Sinusoidal model is found to be the best, since it has highest R2(82.5%), lowest Root Mean Square Error (1229.88), lowest Mean Absolute Error (933.879) for area and highest R2 (78%), lowest Root Mean Square Error (61921), lowest Mean Absolute Error(44665) for production, respectively. Hence, Sinusoidal model is the most suitable fitted model and it is very useful to study about the trend analysis of area and production of banana in Coimbatore.
Key words :
Banana, MAE, Nonlinear models, R2, RMSE.