Sampling

Clicking on the tab “Sampling” will bring forward the sampling pane. The sampling pane is shown in figure 1.

Figure 1: Sampling pane.

Cases can be generated in two different ways: MCAR (Missing Completely at Random) or MAR (Missing at Random). MCAR sets some values to N/A by removing values of some nodes in the generated case set randomly (i.e., MCAR considers all the values in the generated cases and not one case at a time).

MAR randomly sets some values in a case to N/A based on auto-generated templates. These templates specify that if some nodes have a specific value, then the values of a subset of the nodes are randomly set to N/A. Note that the specification of the templates is generated randomly.

In both cases the conditional probability distribution is considered as a factor to the number of cases generated with a specific set of observations. Also in both MCAR and MAR it is possible to indicate the percentage of missing values.

The cases generated will be based on the conditional probability distribution given by the scenario chosen in the combo box “Sample configurations based on”.

Any cases generated will be added to the current set of imported cases.