Modelling of Jassids (Amrasca biguttula) in Cotton: A Count Time Series Approach
B. Venkataviswateja
*
Department of Statistics and Computer Applications, Agricultural College, Bapatla, India.
V. Srinivasa Rao
Department of Statistics and Computer Applications, Agricultural College, Bapatla, India.
A. Dhandapani
Department of Statistics, ICAR-NAARM, Hyderabad, India.
G. Raghunadha Reddy
Department of Economics, AMIC, Regional Agricultural Research Station, Lam, Guntur, India.
D. Ramesh
Department of Statistics and Computer Applications, Agricultural College, Bapatla, India.
A.D.V.S.L.P. Anand Kumar
Department of Entomology, Regional Agricultural Research Station, Maruteru, India.
M. Sivarama Krishna
Department of Entomology, Regional Agricultural Research Station, Nandyal, India.
*Author to whom correspondence should be addressed.
Abstract
This study was aimed to model Jassids population in cotton at Regional Agricultural Research Station (RARS), Nandyal. The secondary standard meteorological weekwise(SMW) data between 2008-2021 was considered based on data availability in the research station. Count time series models and machine learning models are used for modelling the Jassids population dataset Among the models evaluated in the study, the INGARCH-ANN model performed better than the INGARCH, ZIPAR, ZINBAR, and ANN models, according to error comparison metrics (MSE and RMSE). The statistical significance between the models was assessed using the Diebold-Mariano (DM) test. The order of prediction accuracy of the models under consideration is INGARCH-ANN>ANN> ZIPAR >ZINBAR>INGARCH. Overall, the study suggests that employing the Hybrid model could effectively model the jassids population in cotton at RARS, Nandyal.
Keywords: Modelling, ANN, ZIPAR, ZINBAR, INGARCH, MSE, RMSE