Stata

Why choose Stata?

Stata is a comprehensive software that covers all your needs to analyze your data efficiently and accurately: statistics, visualization, data manipulation, automated reports.

The reference in: 

• Econometrics
• Epidemiology
• Survey research
• Behavioral Science
• Biostatistics
• Data science
• Institutional Research

• Medicine
• Political Science
• Education
• Finance, business and marketing
• Public Health
• Public Policy
• Sociology

Data management

Stata offers many features allowing you a total control over your data:

• Import/Export
• Data transformation
• ODBC support
• Combination of datasets,
• Specific data management (cross-sectional data: panel data, time series, longitudinal data, slice data…)

Key statistical features

Stata gives you access to a wide range of statistical features:

Contrast; Pairewise; Comparison and margins; Epidemiology; GMM; Longitudinal data/panel data; Bayesian analysis; Structural Equation Modelling (SEM); Survey processing; Treatment effects; Time series….

Quality graphics

You can easily generate quality graphics for your publications and reports:

Reproducibility, Testing and Certification (FDA)

• Reproducibility between different operating systems and between different editions of Stata
• Tested, verified, and certified commands and functions
• Certified by the FDA for submissions to regulatory agencies

Ease of use

Stata offers a user-friendly interface allowing easy access to all features, even for non- programmers.

StataCorp Technical Support

• As a user of the current edition of Stata (Stata 19), you are entitled to free technical support
• Quick, clear, precise and detailed answers
• A dedicated team of researchers and experts (master, PHD)

Documentation, resource sharing, user community

As a Stata user, you have access to many resources and can count on a very dynamic community:

• 35 manuals with over 18,000 pages
• User forum, Stata researchers’ blog
• Stata Journal: quarterly class A journal
• Stata International Conferences

Cross-platform compatible

Stata is compatible with Windows, Mac and Linux platforms.

An edition that adapts to your needs

Depending on the power of your computers and the volume of data to be processed, Stata offers a range of 3 different editions to suit your needs:

• Stata/BE: Basic edition; for medium-sized datasets
• Stata/SE: Standard edition; for larger datasets
• Stata/MP: Multi-Processors; the fastest edition of Stata (for dual-core, quad-core and multi-core/multi- processor computers); for the largest datasets

Affordable price

Stata is offered as a complete package, without any additional module!

All Stata Versions offer the same features, but here are several purchase options to fit your specific needs, whether you have large datasets to handle or are in need for the fastest computing ever.

Annual and multiyear subscriptions are the easiest way to secure access to the latest features and keep you focus on your research.

StataNow™ (New)

New features straight from development to you

StataNow™ is a continuous-release version of Stata that offers new features as soon as they are ready, without having to wait for the next major release.

With StataNow™, new features are regularly released throughout the current version, from the latest statistical methods to the best in reporting, data visualization, and the user interface. So, you always benefit from the best version of Stata.

All StataNow™ features are rigorously tested and certified, guaranteeing the same level of quality as usual.

See StataNow™ features:
https://www.stata.com/new-in-stata/features/#statanow

How to take advantage of it?

All Stata annual license holders have access to StataNow™, whether as an annual, multi-year or site license. By subscribing to a Stata 19 annual license, you can thus enjoy continuous updates of StataNow™.

Stata Features

Stata includes a wide range of features to meet all your data science needs. Some of the key features are listed below:

What’s new in Stata 19

Stata 19 includes many new enhancements. Here is a zoom at some key features:

Machine Learning via H20

Use the power of machine learning directly within Stata. With the new h2oml suite, you can leverage H2O’s ensemble decision tree methods such as random forest and gradient boosting machine (GBM) to perform classification or regression. Stata supports automatic hyperparameter tuning, cross-validation, SHAP values, partial dependence plots, and integrated predictions, providing a robust and transparent machine learning workflow.

Watch the video

Conditional average treatment effects (CATE)

Go beyond average treatment effects with the new CATE command. Estimate treatment effects at the individual or group level using robust methods such as partialing out and augmented inverse-probability weighting. Choose from flexible modeling approaches including lasso, generalized random forests, and parametric models. Generate predictions, create visualizations, and perform inference, all within a unified and intuitive framework.

Watch the video

High-dimensional fixed effects (HDFE)

Stata 19 allows you to absorb multiple high-dimensional categorical variables in linear regressions, fixed-effects models, and instrumental variables models using the absorb option with areg, xtreg, and ivregress commands. This approach significantly reduces computation time while preserving the accuracy of your model estimates.

Watch the video

Bayesian variable selection for linear regression

With the bayesselect command, identify the most relevant predictors for your linear regression models while accounting for model uncertainty. Access inclusion probabilities, posterior effects, and credible intervals, fully integrated with Stata’s comprehensive suite of Bayesian postestimation tools.

Watch the video

Interval-censored multiple-event Cox model

Use the new stmgintcox command to analyze interval-censored multiple-event data, often encountered in longitudinal studies. Fit models with fixed or time-varying covariates, test for common covariate effects across events, visualize survival curves, and assess model fit for each event individually.

Watch the video

Stata 19 offers other major new features among which:

• Meta-analysis for correlations
• Correlated random-effects (CRE) model
• Panel-data vector autoregressive (VAR) model
• Bayesian bootstrap and replicate weights
• Control-function linear and probit models
• Bayesian quantile regression via asymmetric Laplace likelihood
• Inference robust to weak instruments
• SVAR models via instrumental variables
• Instrumental-variables local-projection IRFs
• Mundlak specification test
• Latent class model-comparison statistics
• Do-file Editor: autocompletion, templates, and more
• PyStata enhancements
• Robust SEs for VAR models
• Half-Cauchy and Rayleigh priors for Bayesian analysis
• Bayesian predictions in user-defined evaluators
• French-language user interface
• And much more…

Find in detail all the new key features of Stata 19 here.

Configuration

Operating systems

Memory and disk space configuration