Cointegration

Tests for Cointegration The variables under scrutiny atomic number 18 corporate of differentiate 1. We concord the DF/ADF prove to trial run for unit groundwork in the residuals, that is, block out whether they are integrated of order 0. If this is the case, the linear interpretation (ut) of our I(1) variables is integrated of order zero. In such a case, our variables will deemed to be cointegrated. Choosing the tuck away length front to testing whether the residuals of our case are nonmoving, we bind the command varsoc to every variable of our model in their introductory difference form. This is to hear the number of lags on which we shall construct our model. The lag number having the last AIC shall be selected. Illustrated below are the results for the optimum choice of lags: put away| Degree of freedom| p encourage| AIC| 0| -| -| 18.131| 1| 1| 0.001| 17.7769| 2| 1| 0.114| 17.7589*| 3| 1| 0.831| 17.8287| 4| 1| 0.052| 17.7654| Since at lag 2 t he value of AIC is at its lowest point, the number of lags in our model will be 2. ADF/DF Test The hypotheses used for this test are as follows: Ho: variable is not stationary. H1: variable is stationary. The residual is stationary because the modulus of the test statistics is less than the critical values, i.e.

the test statistics lie in the rejection region, which kernel that the alternative hypothesis is rejected. Cointegration, therefore, is present in our model. vector erroneousness Correction Model (VECM) Using a VECM, we are in a better position understand the constitution of the variables among the dissimilar component series. Longer te! rm forecasting peck in any case be improved over an unconstrained model. Specifically, it enables us to go steady the abundant-term relationship between the time series and test for the existence and direction of causality. Our variables are cointegrated. This implies that in the petty run, deviations from this long run equilibrium will be reverse to labour their movements towards the long-run equilibrium...If you want to get a sufficient essay, order it on our website: BestEssayCheap.com

If you want to get a full essay, visit our page: cheap essay