This documentation is for a version of the software that is not covered by Standard Support. Select a different version from the version selector in the banner, or access the latest documentation.

Analyzing Data

Advanced Analytics Tasks

This page highlights some frequently used tasks for analyzing data. For documentation of all CAS actions and SAS procedures by product, see SAS Viya Actions and Action Sets by Name and Product and SAS Procedures by Name and Product.

Deep Learning

Model and score networks

Deep Learning Action Set

Deep Neural Network Action Set

Recurrent Neural Networks Action Set

SAS Visual Data Mining and Machine Learning

The NETWORK Procedure

Detect communities in a network

Calculate centrality metrics by community

Data Science Pilot Action Set

Automate some or all of the data mining workflow

Neural Network Action Set

Train a neural network

Factorization Machine Action Set

Create a factorization machine model

Text Mining Action Set

Perform statistical analysis on text data

SAS Forecast Server

HPF Procedure

Automatically forecast transactional data

HPFDIAGNOSE and HPFENGINE Procedures

Create model specifications and fit them to the data

SAS IML

Run a simulation study in parallel

Read and write CAS tables

Optimization and Simulation

Optimization Action Set

Use derivative-free methods to solve an objective with linear constraints

Solve a simple optimization problem that has nonlinear objective

Network Optimization Action Set

Input a graph by using the readGraph action

Solve the linear assignment problem to minimize relay times

OPTMODEL Procedure

Generate a mixed integer linear program to solve a multicommodity network flow model

OPTLP Procedure

Solve linear programs

OPTQP Procedure

Solve quadratic programs

SAS Visual Forecasting

Time Series Processing Action

Forecast a time series or run user-defined code

Stability Monitoring Action Set

Create and score stability monitoring models

Econometrics and ETS

Arima Action

Model time series data with an ARIMA model

Hidden Markov Model Action Set

Model unobservable variables that follow a Markov chain

CNTSELECT Procedure

Create regression models of nonnegative integers

CCDM Procedure

Model the distribution of aggregated loss

Copula Modeling Action Set

Analyze the dependence structure of multivariate distributions

Panel Data Regression Modeling Action Set

Model combined time series and cross-sectional data

Qualitative and Limited Dependent Variable Modeling Action Set

Model limited dependent variables

Frequently Used Actions and Procedures

Aggregate Action

Use the SAS language for CAS to aggregate variable information

Simple Analytics Action Set

Summarize a computed variable in a CAS table and group results

mdSummary Action

Calculate multidimensional summaries of a numeric variable

dtree Action

Use CASL to train a decision tree

GroupBy Action

Group CAS tables using weighting variables

distinct Action

Get the distinct value count for multiple variables

outlier Action

Perform outlier detection and treatment

MEANS Procedure

Compute the descriptive statistics for specified variables

TRANSPOSE Procedure

Transpose selected variables into observations and create a data set for analysis

COMPARE Procedure

Producing a complete report of the differences

CONTENTS Procedure

Create a report with metadata about the table and variables in SAS data set

Copyright © SAS Institute Inc. All Rights Reserved.

Last updated: February 4, 2021