What is autoregression in SAP BI-RA-PA - SAP Predictive Analytics?

SAP Term: autoregression

• Component: BI-RA-PA

• Component Name: SAP Predictive Analytics

• Description: Model that predicts a variable based on its past values.

Smart SAP Assistant

• Key Concepts: 
  Autoregression is a statistical technique used in SAP Predictive Analytics to predict future values of a variable based on its past values. It is a type of linear regression that uses the same independent variable for multiple times in order to predict the future values of the dependent variable. Autoregression is used to identify patterns in time series data and can be used to forecast future values. 
  
  How to use it: 
  In SAP Predictive Analytics, autoregression can be used to forecast future values of a variable by using its past values. To do this, the user must first select the time series data they want to use for the autoregression model. Then, they must select the number of lags (the number of past values) they want to use for the model. Finally, they must select the type of autoregression model they want to use (e.g., ARIMA or SARIMA). 
  
  Tips & Tricks: 
  When using autoregression in SAP Predictive Analytics, it is important to select the right number of lags for the model. Too few lags may lead to inaccurate predictions, while too many lags may lead to overfitting and inaccurate predictions as well. It is also important to select the right type of autoregression model for the data set being used. 
  
  Related Information: 
  Autoregression is closely related to other statistical techniques such as ARIMA and SARIMA models. These models are also used in SAP Predictive Analytics and can be used in conjunction with autoregression models for more accurate predictions. Additionally, autoregression can be combined with other machine learning techniques such as neural networks and decision trees for more accurate predictions.