Predictive Check run mode

The Predictive Check run mode is available for the Phoenix Population Model. When the Predictive Check run option is selected, the model is first fit with 0 iterations (regardless of N Iter in the user interface) in order to generate the data needed to fill the standard results (such as Residuals, Theta, etc.).

Predictive Check is based on the resulting model fit estimates of Theta, sigma, and Omega. Each simulation replicate generates a dataset just like the original dataset. Both eta (individual variation) and epsilon (observation error) are sampled and used to create a population prediction. Predictive Check provides a series of summary statistics (quantiles) of the simulated data to compare with the observed data.

To run Predictive Check on the final estimates from a model fitting (Simple run mode), users should click the Accept All Fixed + Random button in the Parameters > Fixed Effects sub-tab to accept the final parameter estimates as the initial estimates for the Predictive Check.

This run mode provides the option to enter an initial seed value for the random simulation instead of using a random seed. This option is located in the Run Options tab. Each replicate uses a different seed by adding 100 to the starting seed. If the Keep option is checked, then the same starting seed is used for the next run. Otherwise, Phoenix assigns a random seed and predictive check results will vary if the model is run again. The Keep option only prevents a new seed from being created before the next run. The number in the Seed field is retained from the last run.

The Main tab in the Predictive Check mode provides options applying to all the observed variables. Besides this, each observed variable has a separate tab, and the options provided in each tab depend on whether the observable variable is continuous or discontinuous.

Main tab options
Continuous observed variable tab(s) options
Discontinuous observed variable tab(s) options