The Stata News

Stata 15 is here! We are truly excited about all the new features in this release,
and we hope you will be too.

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


Combine endogenous covariates, sample selection, and endogenous treatment in models for continuous, binary, ordered, and censored outcomes.

Learn more »

Latent class analysis (LCA)

Discover and understand the unobserved groupings in your data. Use LCA's model-based classification to find out

  • how many groups you have,
  • who is in those groups, and
  • what makes those groups distinct.

Learn more »

bayes: logistic ...
and 44 more

Type bayes: in front of any of 45 Stata estimation commands to fit a Bayesian regression model.

Learn more »

Markdown & dynamic documents

Type this,
Get this,
  • Create webpages from Stata
  • Intermix text, regressions, results, graphs, etc.
  • See changes in data or commas automatically reflected on webpage

Learn more »

Linearized DSGEs

Write your model in simple algebraic form. Stata does the rest: solve model, estimate parameters, estimate policy and transition matrices (with CIs), estimate and graph IRFs, and perform forecasts.

Learn more »

Finite mixture models (FMMs)

  • 17 estimators and combinations
  • Continuous, binary, count, ordinal, categorical, censored, and truncated outcomes
  • Survival outcomes

Learn more »

Spatial autoregressive models



                  where you are


Learn more »

Interval-censored survival models

Fit any of Stata's six parametric survival models to interval-censored data. All the usual survival features are supported: stratified estimation, robust and clustered SEs, survey data, graphs, and more.

Learn more »

Nonlinear multilevel
mixed-effects models

    When ...
    your science ...
    says ...
    your model ...
    is ...
    nonlinear in its parameters

Learn more »

Mixed logit models: Advanced choice modeling

Do you walk to work, ride a bus, or drive your car? Which of three insurance plans do you buy? Which political party do you vote for?

We make dozens of choices every day. Researchers have access to gaggles of data about those choices. Mixed logit introduces random effects into choice modeling and thereby relaxes the IIA assumption and increases model flexibility.

Learn more »

Nonparametric regression

When you know something matters.

But have no idea how.

Learn more »

Create Word® documents from Stata

  • Automate your reports
  • Write paragraphs and tables to Word documents
  • Embed Stata results and graphs in paragraphs and tables
  • Customize formatting of text, tables, and cells
Create PDFs, too!

Learn more »

Bayesian multilevel models

Small number of groups?
Many hierarchical levels?
Or simply like the graph above?

Consider Bayesian multilevel modeling.

Learn more »

Threshold regression

Your time-series regression may change parameters at some point in time or at multiple points in time. The activity of foraging animals might follow a completely different pattern at temperatures above some threshold. You may not know the value of that threshold. Finding such thresholds and estimating the parameters within the regimes is what threshold regression does.

Learn more »

Panel-data tobit with random coefficients

Stata has long had estimators for random effects (random intercepts) in panel data.

Now you can have random coefficients, too.

Learn more »

Search, browse, and import FRED data

The St. Louis Federal Reserve makes available over 470,000 U.S. and international economic and financial time series. You can now easily search, browse, and import these data.

Learn more »

Multilevel regression for interval-measured outcomes

Incomes are sometimes recorded in groupings, as are people's weights, insect counts, grade-point averages, and hundreds of other measures. Often we have repeated measurements for individuals, or schools, or orchards, etc. So ... we need multilevel regression for interval-measured (interval-censored) outcomes.

Learn more »

Multilevel tobit regression for censored outcomes

  • Left-censoring, right-censoring, both
  • Censoring that varies by observation
  • Make inferences about either the uncensored or the censored outcome
  • Robust and clustered SEs
  • Support for survey data

Learn more »

Multiple-group generalized SEM

Generalized SEM now supports multiple-group analysis. Easily specify groups and test parameter invariance across groups. GSEM models include

  • continuous, binary, ordinal, count, categorical, and even survival outcomes
  • multilevel models

Learn more »

Tests for multiple breaks in time series

  • Cumulative sum (CUSUM) test for parameter stability
    • CUSUM of recursive residuals
    • CUSUM of OLS residuals
  • Plots with CIs

Learn more »

Panel-data cointegration tests

  • Tests
    • Kao
    • Pedroni
    • Westerlund
  • Total of nine variants of tests

Learn more »

Power for cluster randomized designs
Power for linear regression models
Add your own power and sample-size methods
Heteroskedastic linear regression
Poisson models with sample selection
Zero-inflated ordered probit
Bayesian panel-data models
Bayesian sample-selection models
Bayesian survival models
Panel-data interval regression with random coefficents
Panel-data nonlinear models with random coefficients
More in statistics
Stata in Swedish
Improvements to the Do-file Editor
Transparency in graphs
SVG export
Stream random-number generator
Improvements for Java plugins
Stata/MP—Of course many of the new features are highly parallelized.

This is our biggest release yet. Learn more online.
Come see what's new in Stata 15.

Order Upgrade
Register for the 2017 Stata Conference in Baltimore