SAS^® Visual Forecasting

Generate large numbers of reliable forecasts – quickly and automatically.

Three desktop monitors showing SAS Visual Forecasting

Operate more efficiently and effectively by automatically producing a large number of forecasts for a broad range of organizational planning challenges. SAS Visual Forecasting is built on the open SAS^® Viya^™ platform and supports popular open-source and SAS language coding in a single environment.

SAS Visual Forecasting showing time series modeling on desktop monitor

Streamline and automate your forecasting process.

Produce large-scale time series analyses and hierarchical forecasts automatically. Reduced manual intervention means there's less chance of personal biases influencing the forecasting process. It also frees forecast analysts to focus on high-value forecasts or other strategic tasks, rather than spending time building and monitoring forecasting models for every time series.

Plan better for the future.

Manage your organizational planning challenges by generating forecasts on an enterprise scale – quickly, automatically and as accurately as can reasonably be expected. The software delivers results for millions of forecasts at breakthrough speeds, enabling you to plan more efficiently and effectively for the future.

Use your preferred programming language.

Develop models in the open-source language of your choice – Python, Java, R or Lua. You can access tested and trusted SAS time series and forecasting capabilities, and execute models on the new SAS distributed, in-memory platform – all from within a single, governed analytical environment.

SAS^® Visual Forecasting Demo

Mike Gilliland, Product Marketing Manager at SAS, demonstrates how SAS Visual Forecasting provides a resilient, distributed and optimized time series analysis scripting environment for automatic model generation, automatic variable and event selection, and automatic model selection.

Features

Large-scale automatic forecasting. Generate large quantities of statistically based forecasts in a distributed, in-memory environment without the need for human intervention, unless desired.
Scripting language for distributed processing of time series analysis. Run time series analysis and forecasting on an enormous scale, with a scripting language specially designed to support distributed, in-memory time series analysis. The scripting language is also optimized for the machine it is running on, so users don’t have to rewrite code for different machines.
Time series analysis. Execute user-defined programs to convert time-stamped transactional data into a time series format using the TSMODEL procedure.

Time series modeling. Generate forecast models automatically using the TSMODEL procedure.
Hierarchical reconciliation. Models and forecasts each series in a hierarchy individually, then reconciles them at multiple levels in a top-down fashion. You can adjust a forecast at the top level and apportion it to lower levels so the hierarchy maintains consistency, and individual forecasts (by products, locations, etc.) roll up to the top number.

Full Features List

System Requirements