Manish Thukral and Rahul Patil explain how forecast models in SAP Integrated Business Planning can be configured to implement linear regression in forecasting by leveraging available algorithms, custom configuration, and master data models in SAP Integrated Business Planning for demand.
Here are typical scenarios where linear regression forecasting is applicable:
- Agricultural yield for a specific year depends on factors such as rainfall, area of cultivation, and quantity of fertilizer used. Based on linear regression, the relationship between agricultural yield and independent factors such as rainfall, area of cultivation, and quantity of fertilizer used can be determined in terms of linear equation. Based on this the forecasting for agricultural yield in the future can be done.
- The performance of a software system is dependent on factors such as the number of users accessing the system and the number of transactions happening. Based on past data a statistical relationship can be derived for the performance of the system. That can be used to forecast system availability or control number of users accessing the system.
They describe the configuration to be done with a multiple linear regression strategy with one independent variable to behave as a simple linear regression forecast.