Alpha Beta Parameter use

[former_member546432]

Hi All,

I am using Seasonal Model for my Univariate Profile.

I need some information on what is the Alpha, Gamma Parameter and Period Parameter use? What is the value should I maintained?

What is the Period definition?

Could anyone give information on the Alpha Gamma and Period Parameter use and how value range need maintained.

Mainly I need to know the use of Period Parameter.

Thanks,

Best Regards,

Bhaskar

[rajkj · Accepted solution]

Hi Bhaskar,

The following details compliments the information provided by Pawan and Satish.

Smoothing Factor (alpha): It typically represents the weightage given to the recent sales history (actual value). It's value varies from 0 to 1. If it's closer to 1, you give more weight to recent value.

Trend Factor (Beta): This parameter helps to represent the trend in the demand pattern. It's a smoothing factor to trends i.e. closer to 1 gives more weight to the recent trend.

Seasonality Factor(Gamma): This factor helps to adjust the forecast by applying seasonal percentage. It's again a smoothing factor to seasonal index. If the value is closer to 1, more weight will be given to recent seasonal periods.

           Seasonal index of a period = avg sales per period / avg sales of entire horizon

Periods: Number of periods in a season to drive the calculation of seasonal index. Please note the period is specified by the planning buckets profile you enter in the master forecast profile (t.code /SAPAPO/MC96B)

Thanks,

Rajesh

[satish_waghmare3]

Hi Bhaskar,

SAP Help link given by Pawan is definitely helpful. Additionally, Below is information which you shoud know inorder to know usage of α(Alpha), β(Beta), γ(Gamma) and Period.

The most common forecast model is 1st order exponential smoothing. With 1st order exponential smoothing recent data has a bigger influence than data farther in the past. The weights of the recent data increase with the parameter α.

Analogous to the factor α in 1st order exponential smoothing, β weights the influence of recent trends and γ the influence of recent periods.
Typical values for β and γ are 0.3. If there is only a trend pattern in the data history, γ is zero.
The 2nd order exponential smoothing model without seasonal terms is also called Holt model.
In the opposite case – only seasonality, no trend – β is zero and the model is named after Winters.

Forecast Model Type    Periods for Model Initialisation
Constant                                             1
Trend                                                  3
Seasonal                                             one season
Trend & Seasonal                               3 + one season

Hope this will help.

Thanks

Satish

[former_member209769]

Hi Bhaskar,

Unfortunately, the answer is not so simple and you would need to understand some formula:

You could find how these factors would be used in forecast in the following SAP Help link:

http://help.sap.com/saphelp_scm50/helpdata/en/ac/216b83337b11d398290000e8a49608/content.htm

"Periods" means number of periods in a seasonal cycle. e.g. if your business is directly linked to the seasons in a calendar year, then you could use periods = 12, while using 'monthly' buckets in the planning view. So, you would say that seasons repeat after 12 months.

In the SAP help link that I shared earlier, If you click on the link "exponential smoothing", you would see some more info about alpha.

These parameters are statistical terms. If you are interested in knowing the Statistical principles behind forecasting, then you could give your model name and search on google, you would get sufficient information on the particular model and the relevant parameters.

Thanks - Pawan