The forecast strategy determines the method or the techniques that are used to create the forecast. You set the forecast strategy in the forecast profile. You choose from the following strategies.
To create your own strategy, use functional enhancement APOPR001 in transaction CMOD.
Method 
Forecast Strategy 

System Action 
Constant 
Forecast with constant model 
10 
Uses firstorder exponential smoothing. This strategy is identical to strategy 11. See also Constant Model w. FirstOrder Exponential Smoothing. 
Constant 
Firstorder exponential smoothing 
11 
Uses firstorder exponential smoothing. This strategy is identical to strategy 10. See also Constant Model w. FirstOrder Exponential Smoothing. 
Constant 
Constant model with automatic alpha adaptation (firstorder) 
12 
Uses firstorder exponential smoothing and adapts the alpha factor. See also Constant Model w. FirstOrder Exponential Smoothing and Automatic Adaptation of the Alpha Factor. 
Constant 
Moving average 
13 
The system calculates the average of the values in the historical time horizon as defined in the master forecast profile. This average for n periods of history is the forecast result for each period in the forecast horizon; that is, the forecast is the same in every period. See also Moving Average Model. 
Constant 
Weighted moving average 
14 
The system weights each time series value with a weighting factor. For example, you can define the factors so that recent data is weighted more heavily than older data. You define the weighting factor in a diagnosis group. See also Weighted Moving Average Model. 
Trend 
Forecast with trend model 
20 
Uses firstorder exponential smoothing. This strategy is identical to strategy 21. See also Trend/Seasonal Models w. FirstOrder Exp. Smoothing. 
Trend 
Firstorder exponential smoothing 
21 
Uses firstorder exponential smoothing. This strategy is identical to strategy 21. See also Trend/Seasonal Models w. FirstOrder Exp. Smoothing. 
Trend 
Secondorder exponential smoothing 
22 
Uses firstorder exponential smoothing. See also Models with Second Order Exponential Smoothing. 
Trend 
Trend model with automatic alpha adaptation (secondorder) 
23 
Uses secondorder exponential smoothing and adapts the alpha factor. See also Models with SecondOrder Exponential Smoothing and Automatic Adaptation of the Alpha Factor. 
Seasonal 
Forecast with seasonal model 
30 
Uses firstorder exponential smoothing. This strategy is identical to strategy 31. See also Trend/Seasonal Models w. FirstOrder Exp. Smoothing. 
Seasonal 
Seasonal model based on Winters' method 
31 
Uses firstorder exponential smoothing. This strategy is identical to strategy 30. See also Trend/Seasonal Models w. FirstOrder Exp. Smoothing. 
Seasonal 
Seasonal linear regression 
35 
Calculates seasonal indexes, removes the seasonal influence from the data, performs linear regression, and reapplies the seasonal influence to the calculated linear regression line. See also Seasonal Linear Regression. 
Seasonal trend 
Forecast with seasonal trend model 
40 
Uses firstorder exponential smoothing. This strategy is identical to strategy 41. See also Trend/Seasonal Models w. FirstOrder Exp. Smoothing. 
Seasonal trend 
Firstorder exponential smoothing 
41 
Uses firstorder exponential smoothing. This strategy is identical to strategy 40. See also Trend/Seasonal Models w. FirstOrder Exp. Smoothing. 
Automatic model selection 
Forecast with automatic model selection Test for constant, trend, seasonal, and seasonal trend (model selection procedure 1) 
50 
Choose this strategy if you have no knowledge of the patterns in your historical data. The system tests the historical data for constant, trend, seasonal, and seasonal trend patterns. The system applies the model that corresponds most closely to the pattern detected. If no regular pattern is detected, the system runs the forecast as if the data revealed a constant pattern. In this process, the alpha, beta, and gamma factors are determined as follows: · The smoothing factors are taken from the univariate profile. · The settings in the demand planning desktop are used if these are different to the ones in the univariate profile. · If you have made no settings either in the univariate profile or on the demand planning desktop, the default factors of 0.3 are used. See also Automatic Model Selection Procedure 1. 
Automatic model selection 
Test for trend (model selection procedure 1) 
51 
Choose this strategy if you think that there is a trend pattern in your historical data, and if you know that there is no other pattern. The system subjects the historical values to a regression analysis and checks to see whether there is a significant trend pattern. If not, the system runs the forecast as if the data revealed a constant pattern. The alpha and beta factor is also determined as described under strategy 50 for determining the alpha, beta, and gamma factors. 
Automatic model selection 
Test for season (model selection procedure 1) 
52 
Choose this strategy if you think that there is a seasonal pattern in your historical data, and if you know that there is no other pattern. The system clears the historical values of any possible trends and carries out an autocorrelation test. If no seasonal pattern is detected, the system runs the forecast as if the data revealed a constant pattern. The alpha and beta factor is also determined as described under strategy 50 for determining the alpha, beta, and gamma factors. 
Automatic model selection 
Test for trend and season (model selection procedure 1) 
53 
Choose this strategy if you think that there is a seasonal and/or a trend pattern in your historical data. The system subjects the historical values to a regression analysis and checks to see whether there is a significant trend pattern. It also clears the historical values of any possible trends and carries out an autocorrelation test to see whether there is a significant seasonal pattern. If a seasonal and/or trend pattern is detected, a trend model, seasonal model, or seasonal trend model is used. If no regular pattern is detected, the system runs the forecast as if the data revealed a constant pattern. The alpha, beta, and gamma factor is also determined as described under strategy 50. 
Manual model selection with test for an additional pattern 
Seasonal model and test for trend (model selection procedure 1) 
54 
Choose this strategy if you think that there is a trend pattern in your historical data, and if you know that there is a seasonal pattern. The system subjects the historical values to a regression analysis and checks to see whether there is a significant trend pattern. If there is, a seasonal trend model is used. Otherwise, a seasonal model is used. The alpha, beta, and gamma factor is also determined as described under strategy 50. 
Manual model selection with test for an additional pattern 
Trend model and test for seasonal pattern (model selection procedure 1) 
55 
Choose this strategy if you think that there is a seasonal pattern in your historical data, and if you know that there is a trend pattern. The system clears the historical values of any possible trends and carries out an autocorrelation test. If the test is positive, a seasonal trend model is used. Otherwise, a trend model is used. The alpha, beta, and gamma factor is also determined as described under strategy 50. 
Automatic model selection 
Model selection procedure 2 
56 
Choose this strategy if you wish highly detailed tests of the historical data to be carried out. The system tests for constant, trend, seasonal, and seasonal trend patterns, using all possible combinations for the alpha, beta, and gamma smoothing factors, where the factors are varied between 0.1 and 0.5 in intervals of 0.1. The system then chooses the model with the lowest mean absolute deviation (MAD). Procedure 2 is more precise than procedure 1, but takes longer. 
Copy history 
Historical data used as forecast data 
60 
Choose this strategy if demand does not change at all and you want to opt for the least performance or workintensive strategy. No forecast is calculated. Instead, the historical data from the previous year is copied. 
Manual forecast 
Manual forecast 
70 
Choose this strategy if you wish to set the basic value, trend value, and/or seasonal indexes yourself. See also Manual Forecast. 
Croston 
Croston Method 
80 
Choose this method if demand is sporadic. See the APO Glossary and Croston Method. 
Linear regression 
Simple linear regression 
94 
The system calculates a line of best fit for the equation y = a + bx, where a and b are constants. The ordinary least squares method is used. 
On the demand planning desktop, you can see the effects on the forecast results of choosing different methods and techniques by experimenting with the settings under the Model and Parameters tabs of the Forecast view.
Forecast strategies 50 through 55 require a different number of historical values for different tests. For more information, see Model Initialization.