An Empirical Analysis of Stock Market Performance and Economic Growth: Evidence from India

Error Correction Model (ECM)

According to the ECM approach, when two variables, for example Y_t and X_t are cointegrated, there must also be an error correction model that describes the short run dynamics or adjustments of the

cointegrated variables towards their long run equilibrium. EC model consists of last period error as well as lagged values of first differences of each variable. Let us consider, the ECM specification, which can be written as follows:

 y _t

  ₁  y _{t  1}

  ₂  x_t

  z _{t  1}

• 
   _t
  

(3.13)

 x_t

  ₁  x_{t 1}   ₂  y_t

  z_{t 1}   _t

(3.14)

Where, Z_t-1 is an equilibrium error. An important advantage in the ECM is that all the considered variables are stationary and standard OLS is therefore valid. This also includes both the long run and short run information. Where α₂ and β₂ are the impact multiplier (the short run effect) that measures the immediate impact that a change in X_t (or Y_t) will have on a change in Y_t (or X_t). On the other hand,  (or θ) provides us with information about the speed of adjustment in cases of

disequilibrium (long run).

4.