New in Stata 16—lasso, truly reproducible reporting, meta-analysis, frames—multiple datasets in memory, choice models, Python integration, new in Bayesian analysis, panel-data ERMs, imports data from SAS and SPSS, nonparametric series regression, and more.
The Stata News: Vol 34 No 1
 

Dear %%First Name%%,

We always love hearing about what you've been up to. Here's what we've been up to.

Lasso

Model selection and prediction alongside cutting-edge inferential methods.

Truly reproducible reporting

Create customized Word, HTML, PDF, and Excel files with Stata results and graphs.

Meta-analysis

Powerful, yet easy to use, tools for summarizing, evaluating, and testing results from multiple studies.

Frames—multiple datasets in memory

Load datasets simultaneously into frames. Link related frames. Multitask. Record results separately. Make code run faster.

Choice models

Choice models are easy to fit and hard to interpret. Not anymore. Now you can truly interpret your results.

Python integration

Take advantage of any Python package from within Stata. Call machine-learning packages, scrape data, perform image recognition, and more.

New in Bayesian analysis

Multiple chains. Gelman–Rubin convergence diagnostics. Bayesian predictions. Posterior predictive p-values. And more.

Panel-data ERMs

ERMs account for endogenous covariates, sample selection, and treatment all at the same time. And now add panel data to that list.

Import data from SAS and SPSS

Your colleagues give you data in another format. You want to analyze the data in Stata.

Nonparametric series regression

Not sure how to specify your model? Linear or nonlinear? Cubic or quadratic? Try nonparametric series regression.

And that's not all:

Sample-size analysis for CIs

Panel-data mixed logit

xtheckman

Heteroskedastic ordered probit

Point sizes for graphics

Mac Dark Mode

NLMEMs—growth, PK models

Nonlinear DSGE models

Multiple-group IRT

Numerical integration

Linear programming

Stata in Korean

Mac tabbed windows

Do-file Editor autocompletion

 
WHAT'S NEW
 
 
ORDER
 

Kind regards,
Bill Gould, President, and the rest of the Stata development team