Vol. 12 No. 2
Application of INAR model on the pest population dynamics in Agriculture
Author(s): H. S. ROY, R. K. PAUL, L. M. BHAR AND P. ARYA
Abstract:

In real world phenomenon, somernsituations often occur where observations are not continuous. Such situationsrnin time series may be number of road accidents, number of patients hospitalized,rnnumber of phone calls, the number of products sold, etc.  In agriculture, it may be number of pestsrnemerged in a crop season, number of crops infected by a certain type ofrndisease. In these situations, usual autoregressive (AR) model where errors followrnGaussian distribution is not applicable. Hence there is a need to developrnmodels for count observations. The frequently used model for count observationrnis Poisson regression. In this paper, Integer valued autoregressive (INAR)rnmodel is applied for prediction of count observations. Here two real data setsrnare applied where prediction has been done on pest populations. Finally INARrnmodel is compared with Poisson regression and it is seen that INAR model givesrnbetter forecast than the Poisson regression model.

Download Full Article: Click Here