Overview

Predictive Analytics 123 provides a forecast baseline based on historical data patterns. The solution cycles through multiple statistical forecast methods based on seasonality and trend to determine the most accurate forecast. It creates and presents models to support forecasting, prediction, and what-if analysis to determine the most appropriate forecast scenario.

Some key points to know about Predictive Analytics are:

• No use of external variables for predictive enhancement

• Uses past target values (history) to produce future results. Example: Using the last 2 years of monthly total sales values to produce estimates for each of the next 6 months.

NOTE: The more volatile the historical data the more likely the model is to produce negative values in the forecast.

• Includes 12 base algorithms

• Each run has 10+ variations of models

• ARIMA models require 48 months of historical data

• Easy to interpret and understand

IMPORTANT: Predictive Analytics 123 is more effective when run on dense historical data. If you have months in your historical data that are missing a value, you should input a zero in that month.