New features of Stata 16

  • Lasso

    Lasso is a machine-learning technique used for model selection, prediction, and inference.

  • Meta-analysis

    Explore and combine the results from different studies

  • Choice models

    A new, unified suite of commands for modeling choice data.

  • Reproducible reporting

    Reporting features allow you to create Word, PDF, Excel, and HTML documents that incorporate Stata results and graphs with formatted text and tables.

  • Python integration

    Embed and execute Python code from within Stata.​

  • Frames - Multiple datasets in memory

    Work with multiple datasets in memory.  Link datasets and access variables in the datasets in memory.

  • Bayesian - Multiple chains

    Bayesian inference based on an MCMC (Markov chain Monte Carlo) sample is valid only if the Markov chain has converged.

  • Bayesian - Predictions

    These predictions are useful for checking model fit and for predicting out-of-sample observations.

  • Panel data ERMs

    Fit models that account for three common problems that arise in observational data—endogenous covariates, sample selection, and treatment—either alone or in combination.

  • Sample-size analysis for confidence intervals

    The new ciwidth command performs Precision and Sample Size (PrSS) analysis, which is sample-size analysis for confidence intervals (CIs).

  • Import from SAS

    You can now import data stored in SAS (.sas7bdat). Dialog boxes make it easy to explore the data before importing.

  • Import from SPSS

    You can now import data stored in SPSS (.sav). Dialog boxes make it easy to explore the data before importing.

  • Nonparametric series regression

    Fit nonparametric series regressions that approximate the mean of the dependent variable using polynomials, B-splines, or splines of the covariates

  • Nonlinear DSGE models

    The dsgnl command fits nonlinear DSGE models, which means that you no longer need to linearise the models first.

  • Multiple-group IRT models

    The irt commands allow comparisons across groups. Take any of the existing irt commands, add a group(varname) option, and fit the corresponding multiple-group model.

  • Panel-data mixed logit

    Sometimes individuals make the same decision repeatedly.  This is panel data. Relaxes the independence of irrelevant alternatives assumption.

  • xtheckman

    Missing data is sometimes not random.  Heckman selections help fit two level panel data.

  • Nonlinear mixed effects with lags, leads and differences

    New features for fitting nonlinear mixed-effects models (NLMEMs) that may include lag, lead (forward), and difference operators

  • Hetroskedastic ordered probit

    Read more about heteroskedastic ordered probit models in the Stata Base Reference Manual; see [R] hetoprobit.

  • Linear programming

    New Mata class LinearProgram() solves linear programs. It uses Mehrotra's (1992)interior-point method, which is faster for large problems than the traditional simplex method.

  • Numerical integration

    Mata's new Quadrature() class provides adaptive Gaussian quadrature for numerically integrating univariate functions.

  • Point sizes for graphics

    Specify the size of graph elements in printer points, inches, centimetres and relative sizes.

  • Stata in Korean

    All of Stata's interface—all menus and all dialogs—is now available in Korean

  • Mac Dark Mode

    Dark Mode is a color scheme that darkens background windows and controls, so it directs your focus to what you are working on.

  • Improvements in the do-file editor

    The do file editor provides syntax highlighting for Python and Markdown, and now has auto-complete.

New Zealand

+64 9 889 2231

  • White Facebook Icon
  • White LinkedIn Icon
  • White Twitter Icon

SDAS New Zealand

ABN 37 051 831 826

PO Box 5044



Need advice on Stata - Call us on Skype now

© 2018 by SDAS New Zealand