A free update to Stata 13 is available—Stata 13.1
For those who have Stata 13, just type update query in Stata, and follow the instructions, or select “Check for updates” from the Help menu. Stata 13.1 introduces several new features.
Censored outcomes
If you analyze data with Gaussian dependent variables that are censored, you will want to update to Stata 13.1. You can now do just about anything you want with such outcomes. Extensions of tobit and censored regression models include the following:
- Selection models
- Random effects and random coefficients
- Endogenous covariates
- Treatment effects (ATEs)
- Multivariate models
- Unobserved components
- Multilevel models
- Endogenous switching models
All of these models may be combined with each other. For example, you can specify a tobit model with random effects, random coefficients, sample selection, and endogenous covariates. Moreover, the random coefficients can occur in both the outcome and selection models.
All of these features are implemented through extensions to Stata’s gsem (generalized SEM) command and graphical SEM Builder.
Power and sample size
;)
The power command that was introduced in Stata 13 has new methods for analysis of ANOVA models:
- One-way models
- Two-way models
- Repeated-measures models
Like other power methods, you can compute (1) sample size, (2) power, or (3) effect size. Compute any of the three given the other two.
You just tell power what you know, and it produces tables and graphs of what you want to know.
Stata 13.1 also introduces facilities to easily add your own new methods to the power command and produce tables and graphs of results automatically.
Time series
;)
;)
Stata 13.1 also adds several new features for the analysis of univariate time series:
- IRFs (impulse–response functions) for ARIMA and ARFIMA models
- Autocorrelation functions from ARIMA and ARFIMA models
- Parametric spectral densities for seasonal ARIMA models
- Stability checks for ARIMA models
All of your favorite multivariate tools can now be applied to univariate models.
