August 2025

To view this email in your web browser, click here.

Miles and Confidence: Vehicle Types on the Move
Take a look at this horizontal bar chart showing vehicle mileage by type, built with PROC SGPLOT’s HBAR and HBARPARM magic. Bonus: it includes confidence limits to show how sure we are about each number!

Dear Reader,

It's time to mark your calendar for SAS Innovate 2026, the ultimate data and AI event for business leaders, technical users, and SAS partners!

If you thought SAS Innovate 2025 was a blast, the 2026 edition will be even better. We’re heading to Grapevine, Texas from April 27–30, 2026 for another unforgettable experience. Whether you're looking to boost productivity, explore cutting-edge AI, or soak up some serious inspiration from your peers, this is the place to be.

Until then, SAS Innovate On Tour is still rolling through cities around the globe for the rest of 2025. These free, high-energy events bring the same innovation-packed agenda to your backyard, so you can connect, learn, and get inspired without the travel. You might even bump into some SAS Hackathon ambassadors (and don’t forget to register for that… time is running out).

So go ahead, save that date, and get ready to innovate. We’ll see you in Texas!

Thiago De Souza

Editor, SAS Tech Report

Someone forwarded you this edition of Tech Report? Subscribe to receive your own copy each month! Back issues “live” on the SAS Support Community.

Tech tips with Chris

In the SAS Support Community, few things stir up excitement like a file that’s been labeled "complicated." Recently, a user posted a challenge: how to read a quirky, non-standard file into SAS. The file had odd delimiters, inconsistent formatting, and just enough mystery to make it irresistible.

Within hours, the thread lit up with responses. SAS programmers—true DATA step operatives—responded quickly with elegant solutions, clever tricks, and deep knowledge of the INPUT statement. From @ pointer controls to STOPOVER and TRUNCOVER options, the community showcased the unmatched versatility of SAS in reading any file format.

This is where SAS shines. Whether it’s fixed-width, delimited, hierarchical, or just plain weird, SAS has the tools to tame it. The DATA step isn’t just a way to read data—it’s a language for understanding it, reshaping it, and making sense of the messiest inputs. (You can learn these tricks for yourself in this free Ask the Expert training session: The Lost Art of Reading and Loading Raw Data in SAS.)

So next time you're staring down a "complicated" file, remember: it’s not a problem—it’s an invitation. And if you need backup, call in the SAS Community as your own special forces team.

Tips & tricks

SAS Models, Fairness Monitoring | SAS Viya June and July Release
In this month's SAS Viya Release Highlights Show, we focus on ready-made AI models and how SAS is bringing AI directly to your doorstep. Our experts will also discuss the Trustworthy AI Life Cycle and publishing SAS models in Databricks in SAS Model Manager. Plus, hear about some exciting SAS Studio updates including the new data connection manager, code editor, and job enhancements. Additionally, we highlight new features such as the DuckDB Access Engine in SAS Studio, an LLM custom step generator with Azure OpenAI and more!

Cohen's D statistic in SAS
When comparing the means of two groups, it can be useful to standardize the measurements. A popular standardized statistic in the social sciences is called Cohen’s d. In this blog post, Rick Wicklin discusses Cohen’s d statistic and how to implement it in SAS. In a follow-up post, he shows how to obtain confidence intervals, too.

SAS Tutorial | Partitioning Data in Model Studio - for SAS Enterprise Miner Users
When building predictive models, honest assessment on a hold-out sample is the tried-and-true method to best guarantee model generalizability as well as to obtain a fair measure of model performance. In this SAS How To Tutorial, Senior Analytic Training Consultant Jeff Thompson explains the differences and similarities between partitioning data in Model Studio versus SAS Enterprise Miner.

SAS 9.4 News for z/OS (MVS)
SAS is actively testing SAS 9.4M8 and M9 on IBM’s upcoming z/OS Version 3 Release 2 and z17 systems, with plans to support the 64-bit IBM Semeru Runtime Certified Edition for z/OS. These efforts reinforce SAS’s commitment to hybrid cloud and AI readiness for enterprise clients leveraging mainframe platforms.

A case of nostalgia: Clippy once walked so that AI could run
Reflecting on the quirky legacy of Clippy, Jarno Lindqvist explores how modern AI copilots have evolved into context-aware assistants that boost productivity, enhance decision-making, and support developers with intelligent coding tools. It highlights the shift from novelty to necessity as AI becomes an integral part of everyday workflows.

CI/CD for Python and SAS: Build modern workflows with GitHub Actions
Sean Ford shows us how to modernize legacy SAS workflows by integrating Python and automating processes using GitHub Actions and SAS Viya Workbench. This guide walks through building collaborative CI/CD pipelines across diverse development environments, streamlining production and boosting efficiency.

Build without bias: Bias mitigation built-in with ML procedures in SAS Viya
SAS Viya now features built-in bias detection and mitigation across its machine learning procedures, helping users create more ethical and trustworthy AI models. This update reduces manual effort and addresses critical issues like prediction, training data, and intersectional bias to prevent discriminatory outcomes.

Demand planning with SAS
Discover how SAS Demand Planning on Viya 4 empowers retailers to sense demand shifts in real time, optimize forecasts with machine learning, and drive agile supply chain decisions. Learn how intelligent planning tools help deliver the right product, in the right quantity, at the right time.

Building seamless data pipelines across multiple languages: A life sciences and healthcare use case ...
Emmett Smith demonstrates how combining SAS and R in a unified environment streamlines data science workflows in life sciences and healthcare. This reduces friction, boosts collaboration, and enables seamless transitions from data cleaning to model development and app deployment.

Revolutionizing motor insurance: A model for automated claims assessment
SAS is streamlining motor insurance claims by automating the entire process, from uploading accident images to predicting damage and accessing policy details. This integrated solution improves speed, reduces fraud and human error, and enhances the customer experience by combining data management, machine learning, and user-friendly tools.

Use case

Transforming logistics and package delivery with data and AI
French delivery leader Relais Colis transformed its logistics operations using SAS Visual Analytics, enabling real-time tracking, optimized transport flows, and rapid adaptation during the COVID-19 crisis. The shift from spreadsheets to data-driven decision-making helped streamline operations, improve service quality, and unify internal data across teams.

For developers

SAS Bowl LIV, SAS and Generative AI
You can't go anywhere today and not hear/read/see Generative AI. Generative AI refers to a class of artificial intelligence models designed to create new content, such as text, images, music, or code, by learning patterns from existing data. Unlike traditional AI systems that focus on classification or prediction, generative AI can produce original outputs that mimic human creativity. Technologies like large language models (e.g., GPT), diffusion models for image generation, and generative adversarial networks (GANs) are at the forefront of this field. We're featuring Generative AI in the September edition of the SAS Bowl.

Upcoming learning events

(Each is one hour and complimentary unless stated otherwise.)

Ask the Expert: Data Science Meets A/B and Multi-Arm Bandit Testing in SAS® Customer Intelligence 36...
Join SAS' Suneel Grover at 9:00 a.m. ET on Thurs., September 4 to learn how smarter testing yields dynamic, real-time results with SAS Customer Intelligence 360. You will learn:

How to choose the right testing strategy — from classic A/B to adaptive multi-armed bandits.
Ways to integrate SAS CI360 insights into your existing marketing workflows with minimal disruption.
Best practices for using advanced analytics and machine learning to optimize campaigns in real time.
How to measure success and turn test results into actionable business value.

Ask the Expert: How Do I Make a SAS Macro?
Join SAS' Alison Saito at 11:00 a.m. ET on Tues., September 9 to learn how to automate repetitive tasks using macro programming. You will learn about:

Context/background of the SAS macro facility.
Macro variables.
Macro conditional processing.

Ask the Expert: PROC BGLIMM: The Smooth Transition to Bayesian Analysis
Join SAS' Danny Modlin at 10:00 a.m. ET on Thurs., September 18 to learn about the BGLIMM procedure, allowing you to transition your models to the Bayesian realm. You will:

Discover the difference between using PROC MCMC and PROC BGLIMM for Bayesian analysis.
Explore the diagnostic output provided for Bayesian analysis.
Determine how a Bayesian approach to analysis can be helpful.

Ask the Expert: Scaling Advanced Analytics Across the Enterprise With SAS Custom Analytics Framework
Join SAS' David Frede and Priya Sharma at 10:00 a.m. ET on Tues., September 23 to learn how you can put powerful analytics to work for you – without writing code. You will learn how to: