New in Stata 16

Lasso, elastic net, and square-root lasso. There are two parts to our implementation of lasso: prediction and inference.

Lasso has its roots in

  • machine-learning
  • statistics
  • econometrics

Whit this Stata’s new lasso tools you can do Predict outcomes, Characterize groups and patterns in your data, Search over highly nonlinear potential relationships, Perform inference on covariates of interest and more

With Stata’s reporting features you can easily reproduce report in Word, Excel, PDF and HTML. Dynamic document can update reports as data change. Rerun one command or do-file to automatically change full document.

In Stata 16 you can create Word documents from Markdown; easily include headers, footers, and page numbers in Word documents; and convert HTML to Word and Word to PDF.

All of Stata’s new and previously existing reporting features are now documented in a new Stata Reporting Reference Manual

From mulitple studies it combine to new Meta analysis suite.  Use random-effects, fixed-effects, or common-effect meta-analysis to combine individual results and compute overall effect size. Forest plots allow you to visualize the results. With subgroup analysis or meta-regression, you can explore heterogeneity of studies. And you can evaluate publication bias using funnel plots and the trim-and-fill method

Stata 16 introduces a new, unified suite of features for modeling choice data. You able to summarize your choice data, fit models, and interpret the results.

Take advantage of any Python package from within Stata. Call machine-learning packages, scrape data, perform image recognition, and more. Now you can embed and execute Python code within Stata.

Additions for Bayesian analysis:

  • multiple chains 
  • Bayesian predictions
  • Gelman–Rubin convergence diagnostics
  • Posterior summaries of simulated values
  • Posterior predictive p-values
  • MCMC replicates

You can use multiple chains with Bayesian estimation to evaluate MCMC convergence. And you can now evaluate convergence using the Gelman–Rubin convergence diagnostic. With Bayesian predictions, you can check model fit and predict out-of-sample observations.

Extended Regression Models (ERMs).  ERMs were added last release to Stata. They fit models with problems. Now support panel data (also known as longitudinal data or two-level multilevel data).

Import a SAS file or a subset of it from version 7 or higher into Stata

Import an IBM SPSS file or a subset of it (IBM®SPSS® files version 16 or higher and compressed IBM SPSS files version 21 or higher) into Stata.

Nonparametric series regression can select a polynomial, B-spline, or spline function that closely approximates the mean of your outcome. And you can still make inferences. Explore the response surface, estimate population-averaged effects, and obtain tests and confidence intervals.

Stata’s Do-file Editor provided syntax highlighting for Stata. It still does. In Stata 16, it also provides syntax highlighting for Python and Markdown.

And Stata 16’s Do-file Editor has autocompletion. The editor autocompletes words that already exist in the document, autocompletes Stata commands, and autocompletes quotes, parentheses, braces, and brackets.

For more detail about what is new in Stata 16 you can check in here.