One of the most exciting times for us at StataCorp (and hopefully for you as well) is when we get to announce a new version of Stata, full of new features. Now, we hope to experience that feeling with you much more often.

Historically, we have released a new major version of Stata roughly every two years. We will still continue to do that, but most users will now have access to StataNow – a continuous-release Stata. StataNow gives you access to new features now, as soon as they are ready from the development, testing, and documentation groups. The features in StataNow are some of the same features that will also eventually appear in the next major release of Stata. StataNow users will get additional features on a continuous basis throughout the lifetime of a release.

You can read more about StataNow, including how to get it, and you can see its initial set of additional features. But let me tell you a little more about it here.

Many of you create features in Stata that you share with others via your own sites, the SSC archive, and the Stata Journal. And all of you write your own do-files as you perform your analyses in Stata. Knowing this, let me share with you a few technical details about StataNow.

First, StataNow is Stata. To be exact, the current Stata that most of you have is Stata 18.0. StataNow is Stata 18.5 (which we will call StataNow 18.5 from now on). When you are using StataNow, you should start your programs and do-files with version 18.5, just as you previously started them with version 18.0. Why is the version number different? Because StataNow is newer than Stata 18.0, and it is possible something in it will need to be version-controlled differently than in Stata 18. This is no different than when a new release comes out and it has a different version, 16.0, 17.0, 18.0, etc. As always, StataNow is backward compatible, so any programs, do-files, datasets, and so on from earlier versions will work, without changes, in StataNow.

What if we need to version-control something simultaneously in both Stata and StataNow? We would then release Stata 18.1 and StataNow 18.6.

The documentation and help files for Stata 18.0 and StataNow 18.5 are the same. StataNow features are included in them and clearly marked as such.

The dataset format in StataNow is the same as in Stata.

What are the new features in StataNow, and how often will we add features to StataNow? See the current set of new features. There is no set schedule for releasing new features, but we anticipate new features will be released fairly often – several times a year. We will release no new feature before its time, which means that anything released in StataNow is fully official, tested, validated, certified, and documented, just as all the features we put out in a new release of Stata.

When Stata 19 eventually comes out, it will of course include all the features that have come out along the way in StataNow as well as some additional new ones. Users of StataNow will automatically be able to upgrade to Stata 19 — actually, they will upgrade to StataNow 19.5 when Stata 19.0 comes out, and over time StataNow 19.5 will get additional features as soon as they are ready from the Stata elves.

We are excited to be able to give you the new features we add to Stata on a continuous basis, getting them into your hands sooner!

The new table features introduced in Stata 17 and Stata 18 have made it easy to create and customize tables of descriptive statistics, regression results, and more. These powerful features became popular among our users very soon after they were introduced, and we often get questions from users who want to know how to accomplish specific customizations for their tables. To provide our community with more learning resources, we carefully selected some of the questions that we answered frequently, and turned them into a series of example-enriched FAQs: Read more…

The aim of this blog is to describe two novel features introduced in Stata 18 (released in 2023): 1) framesets and 2) alias variables across frames. These features enable Stata to deal with a multiplicity of potentially very large datasets efficiently and conveniently. Framesets allow you to bundle, save on file, and load in memory a set of related frames that hold datasets. Alias variables allow you to access variables in other frames as if they were part of the current frame, with very little memory overhead. Read more…

Are you writing a book featuring Stata programs or output? We’re here to help! We know you want your book to be modern and accurate in all aspects, including any portions that discuss and demonstrate Stata. That’s why we created the Author Support Program—a program that gives you direct access to Stata experts who will review all the Stata-related content in your book to make sure it is accurate, up to date, and reflective of best practices. Read more…

Artificial intelligence (AI) is a popular topic in the media these days, and ChatGPT is, perhaps, the most well-known AI tool. I recently tweeted that I had written a Stata command called chatgpt for myself that runs ChatGPT. I promised to explain how I did it, so here is the explanation. Read more…

In Stata 17, we introduced the new collect suite of commands for creating and customizing tables and the etable command for easily creating and exporting a table of estimation results. Stata 18 offers another new command, dtable, that easily builds and exports a table of descriptive statistics, often called Table 1 in publications. Now generating tables of descriptive statistics for both categorical and continuous variables is easier than ever. It is worth mentioning that the twin commands etable and dtable are both built on the collect framework we introduced in Stata 17, so they share a lot of properties.

In this post, I’ll demonstrate how to create and export simple tables of descriptive statistics and more complex ones that display statistics by group, test for differences across groups, and more. I will also show how you can use the collect suite of commands to further customize the look of your tables and how to include tables created with dtable in complete reports.

Stata 18 is available now. Visit stata.com/new-in-stata to read all about it.

Major new features include

• Bayesian model averaging
• Causal mediation analysis
• Tables of descriptive statistics (Table 1)
• Heterogeneous DID
• Group sequential designs
• Multilevel meta-analysis
• Meta-analysis for prevalence
• Robust inference for linear models
• Wild cluster bootstrap
• Local projections for IRFs
• Model selection for ARIMA and ARFIMA
• Flexible demand systems
• TVCs with interval-censored Cox model
• Lasso for Cox model
• GOF plots for survival models
• RERI
• IV quantile regression
• IV fractional probit model
• Corrected and consistent AICs
• New spline functions
• Alias variables across frames
• Frame sets
• Boost-based regular expressions
• Vectorized numerical integration
• New reporting features
• Do-file Editor enhancements
• Data Editor enhancements
• Graph colors by variable
• All-new graph style

And there is more.

We are excited about the new features and can’t wait for you to try them out!

Stata Press is pleased to announce the release of A Gentle Introduction to Stata, Revised Sixth Edition by Alan C. Acock. If you, or someone you know, are new to Stata, you will want to check out the revised edition of Stata Press’s longtime best seller, which is now available and fully updated for Stata 17. Read more…

Introduction

Some discussions have arisen lately with regard to which standard errors should be used by practitioners in the presence of heteroskedasticity in linear models. The discussion intrigued me, so I took a second look at the existing literature. I provide an overview of theoretical and simulation research that helps us answer this question. I also present simulation results that mimic or expand some of the existing simulation studies. I’ll share the Stata code I used for the simulations in hopes that it might be useful to those that want to explore how the various standard-error estimators perform in situations that are relevant to your research. Read more…

Stata Press is pleased to announce the release of Microeconometrics Using Stata, Second Edition, Volumes I and II, by A. Colin Cameron and Pravin K. Trivedi. This book not only debuted as Kindle’s #1 New Release but also immediately ranked high on Kindle’s competitive best-seller lists in categories such as Statistics, Microeconomics, Econometrics & Statistics, Education Software, Education Statistics, and Mathematical & Statistical. Read more…