Home  /  Products  /  Features  /  Resampling and simulation methods
ORDER STATA

Resampling and simulation methods

Bootstrap sampling and estimation

  • Bootstrap Stata commands
  • Bootstrap community-contributed programs
  • Bootstrap coefficients from most estimation commands by adding an option
  • Bias estimation
  • Clustered data
  • Stratified data
  • Survey data
  • Obtain bootstrap standard errors from most estimation commands by adding an option
  • CI/variance estimators
    • Normal approximation
    • Percentile
    • Bias-corrected (BC)
    • Bias-corrected and accelerated (BCa)
  • Postbootstrapping
    • Tests of linear and nonlinear combinations of estimates
    • Linear and nonlinear combinations of estimates with confidence intervals

Random-number functions

  • Uniform over interval (0,1)
  • Uniform over interval (a,b)
  • Uniform integers over interval [a,b]
  • Standard normal
  • Normal, mean m and std. deviation 1
  • Normal, mean m and std. deviation s
  • Beta
  • Binomial
  • Cauchy
  • Chi-squared
  • Exponential
  • Gamma
  • Hypergeometric
  • Inverse Gaussian
  • Laplace
  • Logistic, mean m and std. deviation π/√3
  • Logistic, mean 0, scale s, std. deviation sπ/√3
  • Logistic, mean m, scale s, std. deviation sπ/√3
  • Negative binomial
  • Poisson
  • Student's t
  • Weibull, shape a, scale b
  • Weibull, shape a, scale b, location g
  • Weibull (proportional hazards), shape a, scale b
  • Weibull (proportional hazards), shape a, scale b, location g

Random-number generators

  • 64-bit Mersenne Twister generator
  • Stream version of 64-bit Mersenne Twister
    • Allow random numbers to be drawn correctly on multiple machines or in multiple processes
    • Each stream is independent of other streams
    • When the stream generator is used, all random-number functions automatically become stream enabled.
  • 32-bit KISS generator

Jackknife estimation

  • Jackknife Stata commands
  • Jackknife community-contributed programs
  • Jackknife coefficients from most estimation commands by adding an option
  • Hold one out
  • Clustered data
  • Survey data
  • Obtain jackknife standard errors from most estimation commands by adding an option
  • Postjackknifing
    • Tests of linear and nonlinear combinations of estimates
    • Linear and nonlinear combinations of estimates with confidence intervals

Monte Carlo simulation

  • Check the performance of statistical methods
  • Simulate Stata commands
  • Simulate community-contributed commands

Permutation tests

  • Stratified data
  • One-sided tests
  • Two-sided tests
  • Enumeration of all possible permutations

Additional resources

See New in Stata 18 to learn about what was added in Stata 18.