Fixing shocks / nonstationary time series via applying the 1st or 2nd diff in EViews

Nonstationary time series

First of all it is important to understand that if the shocks or nonstationary time series are temporary then it can be removed by applying either 1^st or 2^nd diff. The shocks or nonstationary time series which cannot be removed are known as permanent shocks.

We first apply 1^st diff in order to remove the shocks, if the shocks or nonstationary time series are still present in our data then we apply 2^nd diff which will ultimately remove temporary shocks.

1. Open Eviews6.exe

2. Click on File -> Open -> Foreign Data as Workfile…

3. Open the data file “broadband_1 “ by selecting through the path
C:\Program Files\SPSSInc\Statistics17\Samples\English

4. Select MARKET_1 -> OK

5. Click on Quick -> Series Statistics -> Unit Root Test…

6. In Series Name write: MARKET_1, then click OK

7. Select test type as ADF and 1^st Different in “Test for unit root” in Tab -> OK

8. Shocks can also be checked through the table shown below in output section:

If t-statistic value of ADF model is less then test critical values at 1%, 5% and 10%, than this means shocks are not present in the model. Therefore form the above table t-statistic value of ADF model is -4.92 which is less than test critical value at 1%, 5% and 10%, hence we can conclude that shocks have been removed from this model.

Suppose, shocks may not have been removed by the 1^st diff from this model, then we will apply 2^nd diff in a similar way and if then also shocks are present and are not removed, than for sure the shock present in this model are permanent shocks.