Hi Karthik,
Multiple Linear Regression (Causal Analysis)
The general idea of MLR is to find a correlation between the dependent key figure (the one you want to forecast, usually sales history) and other key figures using historical data.
Univariate Forecast Models
Constant Model
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. (Weighted Average)
Trend, Seasonal and Trend & Seasonal Models
For data history which show either a trend or a seasonal pattern the most common forecast model is 2nd order exponential smoothing. (Holt - Linear Regression, Winters -Seasonal Linear Regresion, Holt & Winters, Croston)
Please go through with follwing link for more detail.
http://help.sap.com/saphelp_scm50/helpdata/en/8f/9d6937089c2556e10000009b38f889/frameset.htm
Regards,
Rajnish
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.