Record Details

ARIMA versus trend-agrometeorological wheat yield modelling

KrishiKosh

View Archive Info
 
 
Field Value
 
Title ARIMA versus trend-agrometeorological wheat yield modelling
 
Creator Roy, Ramanath
 
Contributor Verma, Urmil
 
Subject ARIMA, Versus trend-agrometeorological, Wheat, Yield modelling
 
Description Crop yield forecasting is one of the most important aspects of
agricultural statistics system. An efficient crop forecasting infrastructure is a
pre-requisite for information system about food supply, especially
export–import policies, procurement and price-fixation. The present study
has been conducted in Hisar, Ambala, Rohtak and Gurgaon districts of
Haryana for wheat yield forecasting. The study was broadly categorized into
two parts i.e. ARIMA and Trend-agromet wheat yield modelling.
The long time-series data for the period 1959-60 to 1999-2000
have been used for fitting ARIMA models. Trend agromet models have been
developed on the basis of 1978-79 to 1999-2000 time series data on wheat
yield and meteorological data on maximum temperature, minimum
temperature and rainfall for the same period in all the districts. Trend yield
alongwith average maximum temperature, average minimum temperature
and accumulated rainfall integrated over eleven fortnights (Wheat growth
period) have been used as regressors in trend-agromet models.
ARIMA (0,1,1) for Hisar and Ambala, ARIMA (1,1,1) for Rohtak
and ARIMA (1,1,0) for Gurgaon districts have been fitted for yield forecasting
of Rabi wheat 2000-01, 01-02, 02-03, 03-04 and 04-05 in all the four
districts. The deviation of the predicted yield from the Department of
Agriculture and Remote Sensing estimates was very low, favouring the use of
ARIMA models for short-term forecasting.
Secondly, different trend-agromet models have been developed
for wheat yield prediction 2000-01, 01-02, 02-03 and 03-04 in all the four
districts. The relative deviation of predicted yield with DOA/RS estimates are
within tolerable limits advocating the use of trend-agromet models for
obtaining reliable and timely yield forecasts.
Finally, the forecasting performances of both the models have
been compared. The estimates obtained using both the models are found to
be equally good for yield estimation. Realizing the limitation of availability of
meteorological data during the crop growth season, it is emphasized that
ARIMA models are the best alternative to have pre-harvest forecast yield in
the absence of meteorological informations. The ARIMA estimates bear almost
equal accuracy as trend-agromet based yield estimates.
 
Date 2016-11-24T09:55:51Z
2016-11-24T09:55:51Z
2005
 
Type Thesis
 
Identifier http://krishikosh.egranth.ac.in/handle/1/87345
 
Language en
 
Format application/pdf
 
Publisher CCSHAU