Whether your data require a simple weighted adjustment because of differential sampling rates or you have data from a complex multistage survey, Stata's survey features can provide you with correct standard errors and confidence intervals for your inferences. Simply specify the relevant characteristics of your sampling design, such as sampling weights (including weights at multiple stages), clustering (at one, two, or more stages), stratification, and poststratification. After that, most of Stata's estimation commands can adjust their estimates to correct for your sampling design. And much more.
Multilevel mixed-effects models
Whether the groupings in your data arise in a nested fashion (patients nested in clinics and clinics nested in regions) or in a nonnested fashion (regions crossed with occupations), you can fit a multilevel model to account for the lack of independence within these groups. Fit models for continuous, binary, count, ordinal, and survival outcomes. Estimate variances of random intercepts and random coefficients. Compute intraclass correlations. Predict random effects. Estimate relationships that are population averaged over the random effects. And much more.
Take full advantage of the extra information that panel data provide while simultaneously handling the peculiarities of panel data. Study the time-invariant features within each panel, the relationships across panels, and how outcomes of interest change over time. Fit linear models or nonlinear models for binary, count, ordinal, censored, or survival outcomes with fixed-effects, random-effects, or population-averaged estimators. Fit dynamic models or models with endogeneity. And much more.
Structural equation modeling (SEM)
Estimate mediation effects, analyze the relationship between an unobserved latent concept such as depression and the observed variables that measure depression, model a system with many endogenous variables and correlated errors, or fit a model with complex relationships among both latent and observed variables. Fit models with continuous, binary, count, ordinal, fractional, and survival outcomes. Even fit multilevel models with groups of correlated observations such as children within the same schools. Evaluate model fit. Compute indirect and total effects. Fit models by drawing a path diagram or using the straightforward command syntax. And much more.
Linear, binary, and count regressions
Fit classical linear models of the relationship between a continuous outcome, such as weight, and the determinants of weight, such as height, diet, and levels of exercise. If your response is binary (for example, diabetic or not), ordinal (education level), or count (number of children), don't worry. Stata has maximum likelihood estimators—logistic, ordered logistic, Poisson, and many others—that estimate the relationship between such outcomes and their determinants. A vast array of tools is available to analyze such models. Predict outcomes and their confidence intervals. Test equality of parameters or any linear or nonlinear combination of parameters. And much more.
Account for missing data in your sample using multiple imputation. Choose from univariate and multivariate methods to impute missing values in continuous, censored, truncated, binary, ordinal, categorical, and count variables. Then, in a single step, estimate parameters using the imputed datasets, and combine results. Fit a linear model, logit model, Poisson model, hierarchical model, survival model, or one of the many other supported models. Use the mi command, or let the Control Panel interface guide you through your entire MI analysis. And much more.
Adjusted predictions, contrasts, and interactions
Adjusted predictions and contrasts let you analyze the relationships between your outcome variable and your covariates, even when that outcome is binary, count, ordinal, or categorical. Compute adjusted predictions with covariates set to interesting or representative values. Or compute marginal means for each level of a categorical covariate. Make comparisons of the adjusted predictions or marginal means using contrasts. If you have multilevel or panel data and random effects, these effects are automatically integrated out to provide marginal (that is, population-averaged) estimates. After fitting almost any model in Stata, analyze the effect of covariate interactions, and easily create plots to visualize those interactions. And much more.
Analyze duration outcomes—outcomes measuring the time to an event such as failure or death—using Stata's specialized tools for survival analysis. Account for the complications inherent in survival data, such as sometimes not observing the event (censoring), individuals entering the study at differing times (delayed entry), and individuals who are not continuously observed throughout the study (gaps). You can estimate and plot the probability of survival over time. Or model survival as a function of covariates using Cox, Weibull, lognormal, and other regression models. Predict hazard ratios, mean survival time, and survival probabilities. Do you have groups of individuals in your study? Adjust for within-group correlation with a random-effects or shared frailty model. And much more.
Estimate experimental-style causal effects from observational data. With Stata's treatment-effect estimators, we can use a potential-outcomes (counterfactuals) framework to estimate, for instance, the effect of a health education program in schools on teenage smoking. Fit models for continuous, binary, count, fractional, and survival outcomes with binary or multivalued treatments using inverse-probability weighting (IPW), propensity-score matching, nearest-neighbor matching, regression adjustment, or doubly robust estimators. If the assignment to a treatment is not independent of the outcome, you can use an endogenous treatment-effects estimator. And much more.
Handle the statistical challenges inherent to time-series data—autocorrelations, common factors, autoregressive conditional heteroskedasticity, unit roots, cointegration, and much more. Analyze univariate time series using ARIMA, ARFIMA, Markov-switching models, ARCH and GARCH models, and unobserved-components models. Analyze multivariate time series using VAR, structural VAR, VEC, multivariate GARCH, dynamic-factor models, and state-space models. Compute and graph impulse responses. Test for unit roots. And much more.
IRT (item response theory)
Explore the relationship between unobserved latent characteristics such as hospital satisfaction and the probability of responding positively to questionnaire items related to satisfaction. Or explore the relationship between unobserved health and self-reported responses to questions about mobility, independence, and other health-affected activities. IRT can be used to create measures of such unobserved traits or place individuals on a scale measuring the trait. It can also be used to select the best items for measuring a latent trait. IRT models are available for binary, graded, rated, partial-credit, and nominal response items. Visualize the relationships using item characteristic curves, and measure overall test performance using test information functions. And much more.
Fit Bayesian regression models using a Metropolis–Hastings Markov chain Monte Carlo (MCMC) method. You can choose from a variety of supported models or even program your own. Extensive graphical tools are available to check convergence visually. Compute posterior mean estimates and credible intervals for model parameters and functions of model parameters. You can perform both interval- and model-based hypothesis testing. Compare models using Bayes factors. And much more.
As a Stata user for nearly 25 years, I've always appreciated its clean, consistent interface and the peace of mind that comes from StataCorp's rigorous testing and commitment to accuracy. Now with the past few releases, Stata can do virtually anything required by the practicing statistician — it can fit an enormous range of models used throughout the biological and social sciences and has powerful tools for examining and presenting the results of these models. And with -ml- and Mata (Stata's bytecode-compiled, object-oriented, C-like matrix programming language), it's easy to implement new models when necessary. Stata is the one piece of software I couldn't do without.
— Phil Schumm
Senior Statistician and Director of the Research Computing Group
in the Department of Public Health Sciences at the University of Chicago
Intuitive and easy to use.
Once you learn the syntax of one estimator, graphics command, and data management tool, you will effortlessly understand the rest.
Accuracy and reliability.
Stata is extensively and continually tested. Stata's tests produce approximately 4 million lines of output.
One package. No modules.
When you buy Stata, you obtain everything for your statistical, graphical, and data analysis needs. You do not need to buy separate modules or import your data to specialized software.
Write your own Stata programs.
You can easily write your own Stata programs and commands to share with others or to simplify your work using Stata's do-files, ado-files, and matrix-language program, Mata. Moreover, you can benefit from the thousands of Stata user-written programs.
Stata offers 22 volumes with more than 12,000 pages of PDF documentation containing calculation formulas, detailed examples, references to the literature, and in-depth discussions. Stata's documentation is a great place to learn about Stata and the statistics, graphics, or data management tools you are using for your research.
Top-notch technical support.
Stata's technical support is known for their prompt, accurate, detailed, and clear responses. People answering your questions have master's and PhD degrees in relevant areas of research.
Stata's YouTube has over 100 videos with a dedicated playlist of methodologies important to public health professionals. And they are a convenient teaching aid in the classroom.
Stata Press offers books with clear, step-by-step examples that make teaching easier and that enable students to learn and researchers in public health to implement the latest best practices in analysis.