Search
   >> Home >> Products >> Features
Order Stata

Stata features

New in Stata 14—Bayesian analysis • Panel and multilevel survival • Endogenous treatment effects • IRT • Multilevel models with survey data • Markov switching • Unicode • and much more

Linear models

regression  •  censored outcomes  •  endogenous regressors  •  bootstrap, jackknife, and robust and cluster–robust variance  •  instrumental variables  •  three-stage least squares  •  constraints  •  quantile regression  •  GLS  •  more

Longitudinal data/panel data

random and fixed effects with robust standard errors  •  linear mixed models  •  random-effects probit  •  GEE  •  random- and fixed-effects Poisson  •  dynamic panel-data models  •  instrumental variables  •  panel unit-root tests  •  more

Multilevel mixed-effects models

continuous, binary, count, and survival outcomes  •  two-, three-, and higher-level models  •  generalized linear models  •  random-intercepts  •  random-slopes  •  crossed random effects  •  BLUPs of effects and fitted values  •  hierarchical models  •  residual error structures  •  DDF adjustments  •  support for survey data  •  more

Binary, count, and limited outcomes

logistic, probit, tobit  •  Poisson and negative binomial  •  conditional, multinomial, nested, ordered, rank-ordered, and stereotype logistic  •  multinomial probit  •  zero-inflated and left-truncated count models  •  selection models  •  marginal effects  •  more

Generalized linear models (GLMs)

ten link functions  •  user-defined links  •  seven distributions  •  ML and IRLS estimation  •  nine variance estimators  •  seven residuals  •  more

ANOVA/MANOVA

balanced and unbalanced designs  •  factorial, nested, and mixed designs  •  repeated measures  •  marginal means  •  contrasts  •  more

Exact statistics

exact logistic and Poisson regression  •  exact case–control statistics  •  binomial tests  •  Fisher’s exact test for r × c tables  •  more

Tests, predictions, and effects

Wald tests  •  LR tests  •  linear and nonlinear combinations  •  predictions and generalized predictions  •  marginal means  •  least-squares means  •  adjusted means  •  marginal and partial effects  •  forecast models  •  Hausman tests  •  more

Contrasts, pairwise comparisons, and margins

compare means, intercepts, or slopes  •  compare to reference category, adjacent category, grand mean, etc.  •  orthogonal polynomials  •  multiple-comparison adjustments  •  graph estimated means and contrasts  •  interaction plots  •  more

GMM and nonlinear regression

generalized method of moments (GMM)  •  nonlinear regression  •  more

Simple maximum likelihood

specify likelihood using simple expressions  •  no programming required  •  survey data  •  standard, robust, bootstrap, and jackknife SEs  •  matrix estimators  •  more

Programmable maximum likelihood

user-specified functions  •  NR, DFP, BFGS, BHHH  •  OIM, OPG, robust, bootstrap, and jackknife SEs  •  Wald tests  •  survey data  •  numeric or analytic derivatives  •  more

Resampling and simulation methods

bootstrap  •  jackknife  •  Monte Carlo simulation  •  permutation tests  •  more

Time series

ARIMA  •  ARFIMA  •  ARCH/GARCH  •  VAR  •  VECM  •  multivariate GARCH  •  unobserved components model  •  dynamic factors  •  state-space models  •  Markov-switching models  •  business calendars  •  correlograms  •  periodograms  •  forecasts  •  impulse-response functions  •  unit-root tests  •  filters and smoothers  •  rolling and recursive estimation  •  more

Survival analysis

Kaplan–Meier and Nelson–Aalen estimators,  •  Cox regression (frailty)  •  parametric models (frailty, random effects)  •  competing risks  •  hazards  •  time-varying covariates  •  left- and right-censoring, Weibull, exponential, and Gompertz models  •  more

Bayesian analysis

thousands of built-in models  •  univariate and multivariate models  •  linear and nonlinear models  •  continuous, binary, ordinal, and count outcomes  •  continuous univariate, multivariate, and discrete priors  •  add your own models  •  adaptive Metropolis–Hastings sampling  •  Gibbs sampling  •  convergence diagnostics  •  posterior summaries  •  hypothesis testing  •  model comparison  •  more

Power and sample size

power  •  sample size  •  effect size  •  minimum detectable effect  •  means  •  proportions  •  variances  •  correlations  •  ANOVA  •  case–control studies  •  cohort studies  •  contingency tables  •  survival analysis  •  balanced or unbalanced designs  •  results in tables or graphs  •  more

Treatment effects

inverse probability weight (IPW)  •  doubly robust methods  •  propensity score matching  •  regression adjustment  •  covariate matching  •  multilevel treatments  •  endogenous treatments  •  average treatment effects (ATEs)  •  ATEs on the treated (ATETs)  •  potential-outcome means (POMs)  •  continuous, binary, count, fractional, and survival outcomes  •  more

SEM (structural equation modeling)

graphical path diagram builder  •  standardized and unstandardized estimates  •  modification indices  •  direct and indirect effects  •  continuous, binary, count, ordinal, and survival outcomes  •  multilevel models  •  random slopes and intercepts  •  factors scores, empirical Bayes, and other predictions  •  groups and tests of invariance  •  goodness of fit  •  handles MAR data by FIML  •  correlated data  •  survey data  •  more

Multiple imputation

nine univariate imputation methods  •  multivariate normal imputation  •  chained equations  •  explore pattern of missingness  •  manage imputed datasets  •  fit model and pool results  •  transform parameters  •  joint tests of parameter estimates  •  predictions  •  more

Survey methods

multistage designs  •  bootstrap, BRR, jackknife, linearized, and SDR variance estimation  •  poststratification  •  DEFF  •  predictive margins  •  means, proportions, ratios, totals  •  summary tables  •  regression, instrumental variables, probit, Cox regression  •  more

Cluster analysis

hierarchical clustering  •  kmeans and kmedian nonhierarchical clustering  •  dendrograms  •  stopping rules  •  user-extensible analyses  •  more

IRT (item response theory)

binary (1PL, 2PL, 3PL), ordinal, and categorical response models  •  item characteristic curves  •  test characteristic curves  •  item information functions  •  test information functions  •  differential item functioning (DIF)  •  more

Multivariate methods

factor analysis  •  principal components  •  discriminant analysis  •  rotation  •  multidimensional scaling  •  Procrustean analysis  •  correspondence analysis  •  biplots  •  dendrograms  •  user-extensible analyses  •  more

Nonparametric methods

Wilcoxon–Mann–Whitney, Wilcoxon signed ranks, and Kruskal–Wallis tests  •  Spearman and Kendall correlations  •  Kolmogorov–Smirnov tests  •  exact binomial CIs  •  survival data  •  ROC analysis  •  smoothing  •  bootstrapping  •  more

Epidemiology

standardization of rates  •  case–control  •  cohort  •  matched case–control  •  Mantel–Haenszel  •  pharmacokinetics  •  ROC analysis  •  ICD-10  •  more

Data management

data transformations  •  match-merge  •  import/export data  •  ODBC  •  SQL  •  XML  •  by-group processing  •  append files  •  sort  •  row–column transposition  •  labeling  •  saving results  •  more

Graphics

line charts  •  scatterplots  •  bar charts  •  pie charts  •  hi–lo charts  •  contour plots  •  GUI Editor  •  regression diagnostic graphs  •  survival plots  •  nonparametric smoothers  •  distribution Q–Q plots  •  more

Graphical user interface

menus and dialogs for all features  •  Data Editor  •  Variables Manager  •  Graph Editor  •  Project Manager  •  Do-file Editor  •  Clipboard Preview Tool  •  multiple preference sets  •  more

Documentation

23 manuals  •  12,000+ pages  •  seamless navigation  •  thousands of worked examples  •  quick starts  •  methods and formulas  •  references  •  more

Basic statistics

summaries  •  cross-tabulations  •  correlations  •  z and t tests  •  equality-of-variance tests  •  tests of proportions  •  confidence intervals  •  factor variables  •  more

Other statistical methods

kappa measure of interrater agreement  •  Cronbach's alpha  •  stepwise regression  •  tests of normality  •  more

Functions

statistical  •  random-number  •  mathematical  •  string  •  date and time  •  more

Internet capabilities

ability to install new commands  •  web updating  •  web file sharing  •  latest Stata news  •  more

User-written commands

user-written commands for meta-analysis, data management, survival, econometrics, more

Programming features

adding new commands  •  command scripting  •  object-oriented programming  •  menu and dialog-box programming  •  Project Manager  •  plugins  •  more

Matrix programming—Mata

interactive sessions  •  large-scale development projects  •  optimization  •  matrix inversions  •  decompositions  •  eigenvalues and eigenvectors  •  LAPACK engine  •  real and complex numbers  •  string matrices  •  interface to Stata datasets and matrices  •  numerical derivatives  •  object-oriented programming  •  more

Embedded statistical computations

Numerics by Stata

Installation Qualification

IQ report for regulatory agencies such as the FDA  •  installation verification

Accessibility

Section 508 compliance, accessibility for persons with disabilities

Sample session

A sample session of Stata for Mac, Unix, or Windows.
The Stata Blog: Not Elsewhere Classified Find us on Facebook Follow us on Twitter LinkedIn Google+ Watch us on YouTube