The study area selected for this paper is the state of Assam. It is a state in the north eastern region of India. Assam is the gateway of this remote region of the country. The present study is based on secondary data collected from various publications of Directorate of Economics and Statistics, Govt. of Assam. Year wise time series data on area, production and productivity (Average yield/hectare) of banana in Assam have been collected for a period of 15 years from 2003-04 to 2017-18. The growth trends of area, production and productivity of banana in Assam have been studied using compound annual growth rates (CAGR). For estimating the CAGR we have considered the following functional form to be fitted to the collected data set.
Functional forms:
1. Linear function:
y = a +bx
2. Quadratic function: y = a +bX + cX
2
3. Exponential function: y = ab
x
Where,
Y = The area/production/productivity.
X = Time variable in years.
a = Intercept.
b = Slope coefficient.
The functional form having the highest Co-efficient of Determination (
R2) is selected for fitting the trend (
Sharma, 2012). In our analysis we have found that the
R2 values of the exponential function of area, production and productivity of banana in Assam for the period of 2003-04 to 2017-18 are higher than the
R2 values of linear and quadratic functions. Hence the exponential functional form has been selected for fitting trend of area, production and productivity of banana in Assam for this said period. The compound annual growth rate is estimated using the following exponential function as explained below.
y = abx
taking the logarithm on both the sides it takes the linear form
log y = log a + log b
which can be written as
log y = b0 + b1x ..............(i)
Where b1 is the regression coefficient of the linear regression equation (i).
The Compound Annual Growth Rate (CAGR) is calculated as:
CAGR(%) = (Anti log b1 - 1) × 100
Instability analysis in the area, production and productivity of banana in Assam is done by using Cuddy-Della Valle index of instability. John Cuddy and Della Valle developed this index for measuring the instability in time series data (
Cuddy and Della Valle, 1978). Although Coefficient of Variation (C.V) is the simplest measure of instability, there are some limitations in its use in time series data. It over-estimates the level of instability in time-series data which exhibits any trend. Cuddy-Della Valle index is considered as a better measure to evaluate instability in time series data. A low value of this index indicates low instability and vice-versa. The Cuddy-Della Valle index is given by
The ranges of Cuddy - Della Valle Instability Index (
Rakesh Sihmar, 2014) are as stated below:
Low instability = 0 < CDVI< 15
Medium instability = 15<CDVI<30
High instability = CDVI>30
The coefficient of variation (C.V.) is obtained by dividing the standard deviation by the mean and expressed in percentage as
We have observed in this study that more area has been brought under banana cultivation during the last few years in Assam. It has been increased from 42,982 hectare in 2003-04 to 53082 hectare in 2017-18. Similarly, productivity also has increased from 13.837 tonnes/hectare to 17.205 tonnes/hectare during this last fifteen year. Eventually the production of banana is also increased over the years from 594645 tonnes in 2003-04 to 913272 tonnes in 2017-18. An attempt is also made to study the relative contribution of area, productivity and their interaction in increased production of banana in the state. The relative contribution of area, productivity and their interaction in increased production of banana is estimated with the help of the following measures (
Sharma, 2015).
DP = Y0 DA + A0 DY + DA DY .........(ii)
Where,
DP = A
n - A
0
DY = Y
n - Y
0
DP = P
n - P
0
A
0, P
0 and Y
0 represent the area, production and productivity in the base year and A
n, P
n and Y
n represent the corresponding area, production and productivity in the n
th year. The first term, Y
0DA the second term A
0 DY and the third term DADY in the above equation (ii) represents productivity effect, the area effect and the interaction effect respectively. The total change in production can thus be decomposed into three effects
viz. productivity effect, area effect and the interaction effect due to the changes in productivity and area.