Indian Seafood Exports: Issues of Instability, Commodity ...

INDIAN SEAFOOD EXPORTS ISSllES OF INS1AHII.II Y. ('OMMODITY CONCFNTHATION 213
Aliisfex =~,,+fi,fblcc,., t ~ ,f Alngc,, + fi,f~lliif~d~,,, +p,fAli!nfgdp,,
to, i~ltish~r,, +BECM(-I)+ c,
. . , (7)
Whcre A is the difference operator. ECM(-I) is error-correction term lagged by one
period In the cotntegrating regression, for integrating short term dynamics tn the
long-run seafood export functton, This functlon allows to estimate short-run
relationships betwccn Instability index of Indian seafood exports and its
determinants, e,, the error term follows normal ltidependent and identically
distributed (i.i.d) properties. The coefficient 6 measures the response of instability
Index of seafood expons In each prriod from the long-ru11 equiltbrium with the
co~ntegration equation normaliscd on ~yfex. The coefficient 6 represents the
propartton of the d~sequllihrium in iisfex in one period corrected in the next period 6.
is ehpected to have a negative sign and be statistically significant.
The modelling strategy adopted to estimate VECM involvcs three steps
Step I Tesrjor Srarronorrh
Before conducting cointegration tests, it is necessary to establi5h the univariate
time series properties of the variables lo confirm all the variables are non-stationary
and integrated of the same order. This is performed by unit-root test, viz., Augmented
Dickey-Fuller (ADF) test. This test fitids out the order of integration, which is the
minimal number oftimes a series has to be differenced until 11 becomes stationary.
Step 2: Derermrnal~on of Oprimum Lag Lengrh
The cointcgration test is based on vector auto regression and is sensitivc to the
number of lags included in the model, therefore first we should determine the optimal
number of lags used in the cointegration test One way to determine the number of
lags is to select the model with minimum information criterion which are based 011
log-likelihood and penalise the inclusion of additional regressors (Greene, 1993). The
study utilises Akaike ~nformation criterion (AIC) to choose the optimum lag length
Step:3 Coinregrarion Tesr and VECM
The purpose of cointegration test is to determine whether a group of nonstationary
series is cointegrated or not. The presence of cointegration enables to
form a vector error correction mechanism to analyse both the short and long-run
nlationship among cointegrated series.