SAS Visual Forecasting Features List

Large-scale time series analysis & forecasting in a distributed environment

Large-scale time series analysis & forecasting in a distributed environment

  • Automatically generates large quantities of statistically based forecasts in a distributed, in-memory environment.
  • Scripting language enables distributed, in-memory time series analysis.
  • Shuffles the data so that each time series is copied into the memory of a single computing node.
  • Executes each time series on one thread of a node, and each node executes the compiled script for each of its assigned time series.
  • Is optimized for the machine on which it is running so users don’t have to rewrite code for different machines.

Neural network/machine learning modeling strategy nodes

Neural network/machine learning modeling strategy nodes

  • Includes a panel series neural network framework with automatic feature generation and hyperparameter tuning (autotuning) capabilities.
  • Provides a multistage (neural network/regression + time series) framework for creating a forecasting methodology that combines signals from different types of models.
  • Addresses problems that have both time series characteristics and nonlinear relationships between dependent and independent variables using stacked model (neural network + time series) forecasting.

Deep learning capabilities

Deep learning capabilities

  • Produce forecasts with recurrent neural network (RNN), the long short-term memory (LSTM) unit network, and the gated recurrent unit (GRU) network.
  • Transactional data is formatted automatically for forecasting purposes with the above deep learning methods.
  • Recursive strategy is applied automatically for multistep forecasting.

Interactive modeling

Interactive modeling

  • Automatically produce analysis plots, including seasonal cycles, autocorrelation function (ACF), partial autocorrelation function (PACF) and white noise probability test for individual time series.
  • Compare models visually and by using the metric of choice in the in-sample and out-of-sample regions.
  • Develop custom exponential smoothing, ARIMA and subset (factored) ARIMA models for individual time series via a simple user interface.
  • Select your own model champions.

Flexible override facility

Flexible override facility

  • Enables customized forecast adjustments that aren't limited by the structure of the forecasting hierarchy.
  • Lets you select filters based on attributes, such as location, brand, category, size, color, sentiment, quality, etc.
  • Lets you define override specifications by filter and time period(s) for all time series contained within a filter.
  • Includes faceted search filters.
  • Allows disaggregation of override using optimization model.
  • Enables batch execution and incremental data updates.

Integration with open source

Integration with open source

  • Includes External Language Package (EXTLANG), which distributes open source code from Python and R to run in parallel in the worker nodes of SAS Viya in the cloud.
  • Call SAS Visual Forecasting analytical actions from Python, R, Java, JavaScript and Lua.

Hierarchical reconciliation

Hierarchical reconciliation

  • Models and forecasts each series in the hierarchy individually.
  • Reconciles forecasts at multiple levels of the hierarchy.

Automatic segmentation based on data patterns

Automatic segmentation based on data patterns

  • Prebuilt segmentation template based on time series patterns such as volume, volatility, and seasonality.
  • Automatic creation of nested, configurable pipelines with an appropriate modeling strategy for each segment selected by default for the prebuilt demand classification template.
  • Ability to import predefined segments by users, supporting up to 1,000 segments.

Derived attributes

Derived attributes

  • Create predefined sets of derived attributes, including:
    • Time series attributes (min, max, mean, missing, etc.).
    • Forecasting attributes (model properties, statistics of fit).
    • Demand classification attributes.
    • Volume/volatility attributes.

Time series analysis

Time series analysis

  • Autocorrelation analysis.
  • Cross-correlation analysis.
  • Seasonal decomposition and adjustment analysis.
  • Count series analysis.
  • Diagnostic tests for seasonality, stationarity, intermittency and tentative ARMA order selection.

Time frequency analysis

Time frequency analysis

  • Windowing functions.
  • Fourier analysis for real and complex time series.
  • Short-time Fourier analysis.
  • Discrete Hilbert transform.
  • Pseudo Wigner-Ville distribution.

Time series modeling

Time series modeling

  • ARIMA models (dynamic regression and transfer functions).
  • Exponential smoothing models.
  • Unobserved component models.
  • State-space models.
  • Intermittent demand models with Croston’s method.

Automatic time series modeling

Automatic time series modeling

  • Automatic time series model generation.
  • Automatic input variable and event selection.
  • Automatic model selection.
  • Automatic parameter optimization.
  • Automatic forecasting.

Singular spectrum analysis (SSA)

Singular spectrum analysis (SSA)

  • Univariate SSA decomposition and forecasting.
  • Multivariate SSA.
  • Automatic SSA.

Subspace tracking (SST)

Subspace tracking (SST)

  • Perform advanced monitoring (signal analysis) techniques for multiple time series.

Time interval evaluation

Time interval evaluation

  • Evaluate a variable in an input table for suitability as a time ID variable.
  • Assess how well a time interval specification fits date/datetime values or observation numbers used to index a time series.
  • Can either be specified explicitly as input to PROC TSMODEL or inferred by the procedure based on values of the time ID variable.

Time series & forecast viewers

Time series & forecast viewers

  • Provides a Time Series Viewer with a prebuilt set of time series attributes.
  • Provides a Forecast Viewer with a prebuilt set of forecasting attributes.
  • Includes envelope plots for viewing multiple series.
  • Lets you use faceted filters on descriptive statistics, model properties and statistics of fit.

Time series dimension reduction (TDR) package

Time series dimension reduction (TDR) package

  • Enables dimension reduction of transactional time series data in preparation for time series mining.
    • Lets you then apply traditional data mining techniques (such as clustering, classification, decision trees and others).

Project sharing

Project sharing

  • Projects in Model Studio use the project sharing feature of SAS Drive.
  • When shared with read/write access, multiple users can make changes to the project at the same time.
  • Alternatively, projects can be shared with read-only access.

Distributed, accessible & cloud-ready

Distributed, accessible & cloud-ready

  • Runs on SAS® Viya®, a scalable and distributed in-memory engine.
  • Distributes analysis and data tasks across multiple computing nodes.
  • Provides fast, concurrent, multiuser access to data in memory.
  • Includes fault tolerance for high availability.
  • Lets you add the power of SAS analytics to other applications using SAS Viya REST APIs.