Node Definition | General

The General tab of the Node Definition dialog is where the node name and states (depending on the node type) are defined. You can also enter comments about a Node's definition here. If the node is a discrete chance node, you can specify whether the node has separate probability and value inputs. The tab can be accessed by double-clicking on a node in the Influence Diagram or Decision Tree to launch the Node Definition dialog and then selecting the General tab if it isn't already active.

Name and Settings

The Name and Settings section allows you to name the node when it is first created and edit its name subsequent to that. DPL also displays the DPL variable name for the node in this section. For more information on entering node names, see the help topic Node Names.

Alternatives/Outcomes/States (Note: this section title depends on the node type)

The Alternatives/Outcomes/States section allows you to add, insert, delete, and rename states as well as specify a default state. For decision nodes, these state names are alternatives; for discrete chance nodes, they are outcomes; for controlled nodes, they are states. Continuous chance nodes do not have states. The strategies in a Strategy Node are edited on the Data tab. For more information on state names, see Node States.

Dimensions (Note: this section appears only for value nodes)

Value nodes in the Influence Diagram can be a single value (i.e., a scalar), a one dimensional array, a series, or a two-dimensional array. This section allows you to specify the dimensionality of the value node. If you specify anything other than a scalar, the number of rows, columns or both is specified on the Data tab.

Separate Probability and Value Data Input Trees (Note: this box appears only for discrete chance nodes)

With this option turned a chance node's data Data Input Tree will be split in two:one for probabilities and one for values. This is particularly helpful if the values and probabilities of a chance node are not conditioned by the same node.

Hide

To hide the node from view in the Influence Diagram, check this box. To un-hide the node, make sure nothing is selected in modeling window and click Influence Diagram/Decision Tree | Object | Edit to bring up the Select Node dialog. Hidden nodes will be indicated with "**". To edit a hidden node, select it from the list and click OK. This will bring up the General tab of the Node Definition dialog for the node. Un-check the Hide box to un-hide the node.

ConstantNote: this box appears only for value nodes)

Check this box to indicate that the value is a constant not a variable and won't change or be re-calculated.

StringNote: this box appears only for value nodes)

Checking this box indicates that the node will consist of a text string.

DummyNote: this box appears only for value nodes)

Check this box to specify that the node contains no data and will be ignored by DPL during an analysis. See the help topic dummy nodes for more information.

IntegerNote: this box appears only for value nodes)

If the value node is conditioned by a chance node which has the Don't Gamble specification, DPL will make sure the value of the value node is an integer. This is important if the value node is being used to send a flag to the spreadsheet.

Versions: DPL Professional, DPL Enterprise, DPL Portfolio

See Also

Node Names

Node States

Chance and Decision Node Defaults

Influence Diagram Nodes

Decision Tree Nodes

Node Definition

Node Definition | Data

Node Definition | Links

Node Definition | Tree Instance