Contents
- Index
Simulations
The capability to simulate data is available in Program MARK. Three simulation systems are available: Program RELEASE simulations, Program CAPTURE simulations, and simulation of models developed in Program MARK.
A fairly complex set of interactive screens are provided to set up simulations. You must specify the True Model, the values of the beta parameters to use to generate data with the true model (Beta Values), the number of animals released (Releases), 1 or more models to estimate (Estimation Models), and the parameters of the simulation (Simulation Specification). If individual covariates are included in the model, you must also specify the distribution of the Individual Covariates. The capability to bootstrap data is also available, in which case you do not need to specify the True Model, the beta parameter values, or the number of animals released, because the data provide this information.
A series of Tab Windows are provided to set up simulations. The first Tab Window is the Status Tab Window that tells you what tasks you have to complete to perform the simulation. Each task is accomplished by clicking on one of the Tabs at the top of the dialog, filling in the necessary information, and clicking the OK button to accept the contents of the window. When all the necessary tasks have been completed, the Start Simulations button will be enabled, allowing you to proceed.
The simulation job is started in a separate thread, so that it does not conflict with further models you may want to construct. A window is opened that shows the progress of the simulation. For each of the models that are estimated, a case is added to the simulation output database that contains the model name and the quantities checked in the Simulation Specification Tab Window.
Once the simulation job is completed, you can view the results in the View Simulation Results menu choice under the Simulation menu. Simple summaries can be performed with this procedure. However, more sophisticated summaries will require importing the DBF file into Excel or a statistical package such as SAS to perform the anayses.