The steps of exponential smoothing in eviews are as follows:

1, use the command mode: smooth y to get a dialog box, select the exponential smoothing form you want,

2. Then fill in the smoothing parameters in the three options of alpha, beta and gamma respectively. α generally takes a value greater than 0.5 (because it has a large weight in the recent forecast), β and γ generally take 0, and click OK.

3. There are mean and trend terms at the bottom of the predicted results, which have the following relationship: F(t)=trend+mean*t, so that t= 1 can get the predicted value F( 1) for next year, and so on.

Note: The biggest difficulty of this method is how to determine the value of α, but one criterion is to choose different values to minimize the sum of squares of residuals. You can try to determine a more accurate value by dichotomy.