Statistiacl Investigation and Interpretataion of replacement series intercropping experiments with mixtures methodology
Loading...
Date
2010
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
UAS, Dharwad
Abstract
Intercropping has been traditional practice in our country. In recent years research into
intercropping has attracted attention of the agricultural scientists. Therefore this study has
been undertaken to demonstrate the applicability of the mixture experiment in agricultural
research. The object of the mixtures methodology is to obtain individual parameters with an
interpretation closer to that of parameters in ordinary polynomial response functions. In the
methodology used, a relationship is built between crop response and the proportions of crops.
Statistical techniques used to analyze data from Mixture Experiments involve fitting Multiple
Regression models with the intercept set to zero. One purpose of statistical modeling in a
Mixture Experiment is to model the mixing components such that predictions of the response
for any mixture component, singly or in combination, can be made empirically. A chickpea
based mustard intercropping experiment conducted in Agricultural Research Station (ARS),
Gulbarga during the year 2007 and 2008 laid out in RCBD was considered for the study. It is
a replacement series intercropping experiment with 3 replications comprising 5 different row
proportion treatments along with 2 sole crop treatments and satisfies all the criteria to apply
mixtures methodology. About 11 different price combinations were worked out keeping in
mind the small and large variations of prices in the market. The results of multiple
regressions for all the price ratios showed 6:2 row proportion of chickpea + mustard to be
more stable with fluctuations in market prices and proved to be optimum with existing market
price when compared with results of usual ANOVA. Mixtures methodology is proved to be
advantageous over results of ANOVA in aspects like change in optimum area to be
recommended with change in market price, narrow range of confidence interval, prediction of
the optimum response of the row ratio not included in the experiment.
Description
Keywords
Agricultural Statistics