Comparative study of ARIMA and ARIMAX models for sugarcane yield forecasting in Northern Agro-climate zone of Haryana
Loading...
Date
2019
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
CCSHAU
Abstract
Crop yield. forecasting is. one of the most important aspects of agricultural statistics system.
An efficient crop forecasting is pre-requisite for information system about food supply, especially
export–import policies, procurement and price-fixation. The study has been performed in two parts i.e.
the development of ARIMA and ARIMAX models for sugarcane yield prediction in Karnal, Ambala
and Kurukshetra districts of Haryana. The ARIMA models have been fitted using sugarcane yield data
for the period 1966-67 to 2011-12 of Karnal and Ambala districts and 1972-73 to 2011-12 of
Kurukshetra district. However, the fortnightly weather data from 1978-79 to 2011-12 have been
utilized as input for ARIMAX model building. The validity of fitted models have been checked for
subsequent years i.e. 2012-13 to 2016-17, not included in the development of the models.
ARIMA(0,1,1) for Karnal, Ambala and Kurukshetra district have been fitted for sugarcane yield
forecasting. Secondly, the ARIMA models with alternative combinations of weather variables were
tried for fitting ARIMAX models. Consequently, ARIMA(0,1,1) for Karnal, Ambala and Kurukshetra
districts along with fortnightly weather variables (viz., tmx7, tmn1, tmn10, arf8, arf11 over the crop
growth period) as input series were utilized in fitting ARIMAX models.
ARIMA and ARIMAX models both provided the relationships to reliably estimate the
sugarcane yield. The predictive performance(s) of the contending models were observed in terms of
the percent deviations of sugarcane yield forecasts in relation to the observed yield(s) and root mean
square error(s). The level of accuracy achieved by ARIMA model(s) with weather as input was
considered adequate for predicting sugarcane yield(s) i.e. the ARIMAX models consistently showed
superiority over ARIMA models in capturing lower percent deviations pertaining to sugarcane yield
estimates. The ARIMAX models performed well with lower error metrics as compared to ARIMA
models in all time regimes. Five-steps ahead estimated values of sugarcane yield(s) favour the use of
ARIMAX models to get pre-harvest sugarcane yield prediction in the districts under consideration.
Description
Keywords
null