Previous: What larify Does

Up: Clarify: Software for Interpreting

Next: A Simple Example

What's new in ${\mathfrak{C}}$larify 2.0?

${\mathfrak{C}}$larify 2.0 includes a number of enhancements over previous versions, including:

• Support for more models, including weibull regression, seemingly unrelated regression equations, and the additive logistic normal model for compositional data.

• The ability to apply standard transformations - such as natural logs and exponents - to dependent variables, estimate a model, and then reverse those transformations when interpreting the results.

• ${\mathfrak{C}}$larify is Amelia-compatible: if you use multiple imputation to correct for problems with missing values, ${\mathfrak{C}}$larify will analyze all the multiply imputed data sets and appropriately combine the results and compute your quantity of interest automatically. (For information on the software program Amelia, see http://GKing.Harvard.Edu).

• The option to generate antithetical simulations, which guarantees that the mean of the simulated parameters equals the vector of point estimates, $\hat\gamma$, and reduces Monte Carlo variance.

• More powerful commands for setting the values of the explanatory variables (the $X$'s), using either single or multiply-imputed datasets to compute descriptive statistics.

• The ability to re-display the previous point estimates by entering the estsimp command without any arguments.

• The option in the estsimp command to drop previously simulated parameters.