Edit Structural Coefficient

Context

BayesiaLab uses proprietary score-based learning algorithms, which use the Minimum Description Length Score (MDL) to measure the quality of candidate networks with respect to the available data. In BayesiaLab, the computation of the MDL Score also includes the so-called Structural Coefficient α as a weighting factor for the structural component $DL(B)$. Please see Key Concepts: Structural Coefficient for more details.

Usage

There are multiple ways to open the Structural Coefficient window:

• Select Menus > Edit > Edit Structural Coefficient.

• Select Graph Panel Contextual Menu > Structural Coefficient.

• If the Structural Coefficient indicator is displayed in the Status Bar, click on the icon . In the Structural Coefficient window, you can set the desired level in one of three ways:

• Use the slider to select the desired level.

• Use the up/down buttons to change the display value in 0.05 increments.

• Type in the desired value.

Workflow Illustration

To illustrate the usage of the Structural Coefficient and its impact on learning, we use an example presented in the webinar, Analyzing Capital Flows of Exchange-Traded Funds.