The double exponential smoothing procedure draws a graph containing the observations, predicted values, and forecasts, versus time.

 Notice that the fitted value pattern lags behind the data pattern. This lag is because the fitted values are the weighted averages from the previous time unit.

The double exponential smoothing procedure also displays, along with the graph, the smoothing constants and three measures to help you determine the accuracy of the fitted values: MAPE, MAD, and MSD.

·    The smoothing constants (alpha and gamma) are the weights used in computing the components of the fitted value

·    The three measures of accuracy are not very informative by themselves, but they are used to compare the fits obtained by using different methods. For all three measures, smaller values generally indicate a better fitting model.

For the sales data, you should fit other models and compare the accuracy measures before deciding on a final model.