Agrometeorological yield forecasting using a multiple regression a
lways starts with a
table of data containing yields and a s
eries of agrometeorological and other variables
which are thought to determine the yields. An e
xample of such a table is given below
(Illustration 8) with data from Malawi. Such tables are often referred to as the “calibration
matrix.”
Aregression equation (usually linear) is derived between crop yield and one or more
agrometeorological variables, for instance
Yield=5+0.03Rain
March
−0.10T
C,June
with yield in tons ha
-1
, March rainfall in mm and June temperature in °C. Beyond its
simplicity, the main advantages of the equation are the fact that (1) calculations can be
done manually, (2) data r
equirements are limited a
nd (3) the ease of derivation of the
equations using standard statistical packages or a spreadsheet.