Is The Cointergration Approach And The Error Correction Model?

Equation (3) captures the partial adjustment of the model. The equation shows how area cultivated is adjusted based on the difference between actual area cultivated and the previous area cultivated. The above model is the acreage response model and the nerlovian output supply response is also constructed in the same way.

Nerlovian approach has been widely used to the production behavior of agricultural producers. However, Nerlove’s framework is criticized of its weaknesses. First, the model applies the OLS without confirming the stationarity of variables hence the suspicion of non stationarity which may produce spurious results. Second, the model fails to give distinct short and long run estimates. Furthermore, expected price changes based

This happens because generally, economic variables including agricultural time series data are non-stationary. To overcome the problem, cointegration is used since it is used with non-stationary data. Cointegration tests for the existence of long run relationship among variables. The concept of cointegration suggests that if two (or more) series are associated to create an equilibrium relationship over the long run, then although the series may be non-stationary, they tend to move closely together over time and the variation between them is unchanged, that is, stationary ( Soontaranurak, 2011) . Here, a meaningful regression on the levels of the variables can be obtained, and any important long-run information is still present. When there is a long rung relationship among variables, an Error Correction Model (ECM) is estimated. This model gives the speed of adjustment of variables to their long run equilibrium. The ECM gives information on both the short and long-run elasticities. The coefficient of the error correction term is a short term adjustment coefficient, that is, it represents the amount by which the long run disequilibrium is adjusted to