Forecast Models 

When analyzing a time series various patterns can be recognized. The following different forecast models are derived from these patterns:

·        Constant – There is a constant time series gradient when the time series varies statistically around a mean value:

·        Trend – With a time series gradient of type trend the time series value falls or rises constantly over a long period of time with only occasional deviations:

·        Season – With a seasonal time series gradient, recurring variances from a base value occur at periodic intervals:

·        Trend-seasonal – With a trend-seasonal time series gradient, seasonal variances occur from a continually increasing mean:

·        Copying actual data (no forecast is executed) – copies the historic data updated from the operational application. You can then edit this as required.

·        Irregular – No pattern can be detected in a series of consumption values.