Order Stata
## Binary, fractional, count, and limited outcomes

Is your response binary (for example, employed or unemployed), ordinal
(education level), count (number of children), or censored (ticket sales in an
existing venue)? Stata has maximum likelihood estimators—logistic, probit,
ordered probit, multinomial logit, Poisson, tobit, 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.

- Basic (dichotomous) ML logistic regression with influence statistics
- Fit diagnostics and ROC curve
- Classification table and sensitivity-versus-specificity graph
- Skewed logistic regression
- Grouped-data logistic regression
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints
- Multiple imputation

Watch Logistic regression tutorials

**Conditional logistic regression**

- Conditional fixed-effects logit models (m:k matching) with exact likelihood (no limit on panel size)
- For matched case–control groups
- McFadden’s choice model Updated
- 1:1 and 1:k matching
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints
- Predictions for influence and lack-of-fit statistics and Pearson residuals

**Multinomial logistic regression**

**Probit regression**

- Dichotomous outcome with ML estimates
- Bivariate probit regression
- Endogenous regressors
- Grouped-data probit regression
- Heteroskedastic probit regression
- Rank-ordered with alternative-specific and case-specific variables Updated
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints

- Beta regression
- Fractional probit regression
- Fractional logistic regression
- Heteroskedastic fractional probit regression

- Ordered logistic (proportional-odds model)
- Ordered probit
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints

- Lower and upper limits of censoring
- Predictions available for expected value, conditional expected value, censored expected value, and probability of censoring
- Endogenous regressors
- Selection models
- Random effects and random coefficients
- Treatment effects (ATEs)
- Multivariate models
- Unobserved components
- Endogenous switching models
- Robust, cluster–robust, bootstrap, and jackknife standard errors

- Lower and upper limits of censoring
- Differing limits for each observation
- Predictions available for expected value, conditional expected value, censored expected value, and probability of censoring
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints

- Open and closed intervals
- Endogenous regressors
- Selection models
- Random effects and random coefficients
- Treatment effects (ATEs)
- Multivariate models
- Unobserved components
- Endogenous switching models
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints

**Poisson and negative binomial regression **

- Predict expected counts, incidence rates, and probabilities of counts
- Poisson goodness-of-fit tests
- Poisson model with endogenous regressors
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints

**Censored Poisson regression** New

- Left, right, and interval censoring
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints

Watch Censored Poisson regression.

**Zero-inflated count models**

- Zero-inflated Poisson
- Zero-inflated negative binomial
- Predict expected counts, incidence rates, and probabilities of counts
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints

**Left-truncated count models**

- Random-utilities maximization model
- Full maximum-likelihood estimation
- Up to eight nested levels
- Facilities to set up the data and display the tree structure
- Predictions available for utility functions, probabilities, conditional probabilities, and inclusive values
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints, including constraints on inclusive-value parameters

- Alternative-specific and case-specific variables Updated
- Homo- or heteroskedastic variances
- Various correlation structures, including user-specified
- Probabilities based on GHK simulator
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints

- Two-step and maximum likelihood (ML)
- Robust, cluster–robust, bootstrap, and jackknife standard errors (ML only)
- Bootstrap and jackknife standard errors (two-step)
- Linear constraints (ML only)
- Predictions available for Mills’ ratio, expected value, conditional expected value, probability of selection, nonselection hazard, and more

**Heckman selection with a binary outcome **

- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints
- Predictions available for probability of binary outcome, all four combinations of outcome and selection, probability of selection, conditional probability of outcome, and more

**Heckman selection for ordered probit **

- Robust, cluster-robust, bootstrap, and jackknife standard errors
- Linear constraints
- Predictions available for marginal and bivariate probabilities, probabilities of levels conditional on selection or no selection, selection probability, linear production, and more

**Rank-ordered logistic regression **

- Plackett–Luce model, exploded logit, choice-based conjoint analysis
- Complete rankings of ordered outcome
- Incomplete rankings of ordered outcome
- Ties (“indifference”)
- Prediction of probability that alternatives are ranked first
- Robust, cluster–robust, bootstrap, and jackknife standard errors

**Stereotype logistic regression **

- Predictions of probabilities of outcomes
- Robust, cluster–robust, bootstrap, and jackknife standard errors
- Linear constraints

- View and run all postestimation features for your command
- Automatically updated as estimation commands are run

Watch Postestimation Selector.

- Automatically create indicators based on categorical variables
- Form interactions among discrete and continuous variables
- Include polynomial terms
- Perform contrasts of categories/levels

- Estimated marginal means
- Marginal and partial effects
- Average marginal and partial effects
- Least-squares means
- Predictive margins
- Adjusted predictions, means, and effects
- Works with multiple outcomes simultaneously New
- Contrasts of margins
- Pairwise comparisons of margins
- Profile plots
- Graphs of margins and marginal effects

Watch **Introduction to margins in Stata** tutorials

Watch **Profile plots and interaction plots in Stata** tutorials

A single categorical variable

A single continuous variable

Interactions of categorical variables

Interactions of categorical and continuous variables

Interactions of two continuous variables

A single continuous variable

Interactions of categorical variables

Interactions of categorical and continuous variables

Interactions of two continuous variables

Additional resource

*Regression Models for Categorical Dependent Variables Using Stata, Third Edition*by J. Scott Long and Jeremy Freese

See tests, predictions, and effects.

See
**New in Stata 14**
for more about what was added in Stata 14.