The Statistics Forecasting Module allows analysts to foresee trends and develop forecasts, without necessarily being experts in the sector.

Novice forecasting users can create advanced forecasts that consider several variables while more experienced users can use Statistics Forecasting to validate their models.

​

Thanks to Statistics Forecasting it is possible to:

​​

• Develop predictions reliable, regardless of the size of the dataset or the number of variables involved.

• Reduce the error forecasting by automating the selection of the appropriate models and their parameters.

• Update and manage prediction models, so you can focus on exploring why some models deviate from the norm.

 

Statistics Forecasting offers:

​

• Techniques advanced stats required to work with time-series data, regardless of experience level.

• Procedures that they simplify the optimization of the temporal analysis.

 

Advanced Statistical Techniques:

 

• Analyze historical data, predict trends faster, and provide insights that are easy to understand and use for business decision makers.

• Automatically determine the best ARIMA or exponential model for analyzing historical data.

 

Procedures

 

• TSMODEL—Use Expert Modeler to model a series of time variables, using ARIMA or exponential optimization techniques.

• TSAPPLY—apply saved templates to new or updated data.

• SEASON—estimate additive or multiplicative seasonal factors for periodic time series.

• SPECTRA—decomposing a time series into harmonic components, consisting of regular periodic functions with different periods or wavelengths.

• Model thousands of different time series at once, instead of one variable at a time.

• Save models in a central file to update predictions if data changes, without re-setting parameters or recalculating models.

• Write scripts to automatically update models with new data.