Node Description

Matrix Distributions

The Matrix Distributions node takes an Input Correlation Matrix and uses the correlation values between each row and column to create a set of correlated Customer Distributions. Additional Customer Distributions from the Input Attributes List can also be added to the Output Customer Distributions Matrix.

Each output Customer Distribution column will have a Mean and Standard Deviation (SD) set according to either the Configuration Dialog, or overridden by the ‘Mean’ and ‘SD’ columns in the Input Attribute List. By default, Unit Distributions having a Mean of 0.0 and a Standard Deviation of 1.0 will be generated. Each row in the set of Output Customer Distributions corresponds to a part-worth value for a Virtual Customer.

An upstream Differentiation Horizontal node, or Correlation Pairs to Matrix node, or Similarity Matrix node can produce an Output Product Correlation Matrix. This can be treated as the Input Correlation Matrix to this Matrix Distributions node.

The Output Customer Distributions from this Matrix Distributions node can become part of a Customer Willingness To Pay Matrix (WTP Matrix) for a set of Products. The Input WTP Matrix can feed a downstream Market Simulation node or a Market Tuning node.

Both the Input Attribute List and the Input Correlation Matrix is optional. If no inputs are provided, then the Matrix Distributions node will generate a single Customer Distribution with a Mean and SD set according to the Configuration Dialog.

This Community Node documentation assumes you have already downloaded the open-source KNIME analytics platform and installed the free Market Simulation (Community Edition) plugin. If not, start by returning to Getting Started.

Downloads

KNIME Workflow
Workflow Image

#1 No Input

The Input Attribute List (top-input) and the Input Correlation Matrix (bottom-input) are both optional. This workflow shows the data that is generated when no input has been provided.

Inputs

No Input

Both the Input Attribute List (top-input) and the Input Correlation Matrix (bottom-input) is optional. Neither needs to be provided.

Node

Configuration

The Matrix Distribution node only generates Normal (Gaussian) Distributions. The Default Mean and Default Standard Deviation (SD) are used when no Input Attribute List has been provided.

Outputs

Normal Distribution

A single Customer Distribution is created according to the default configurations.

Statistics

The Statistics node shows the Mean, SD, and Histogram.

#2 Input Attribute List

Shows the generated output when the Input Attribute List (top-input) is provided. The Input Attribute List is the set of Products, Features, or other Attributes to add to the Output Customer Distributions Matrix.

Inputs

Attribute List

Customer Distributions are generated for each row in the Input Attribute List. Missing Mean and Standard Deviation (SD) values are overridden by the defaults in the Configuration Dialog.

Node

Configuration

Default Mean and Standard Deviation (SD) override missing values.

Outputs

Output Attributes

Corresponds to the Input Attribute List with missing Mean and SD values filled in.

Output Distributions

One Customer Distribution column for each Attribute row.

#3 Input Correlation Matrix

Shows the generated output when the Input Correlation Matrix (bottom-input) is provided. The Input Correlation Matrix is the set of correlations that define the relationship between the Output Customer Distributions.

Inputs

Correlation Matrix

The set of correlations that define the relationships between the Output Customer Distributions.

Node

Configuration

The Mean and Standard Deviation (SD) are used when when no Input Attribute List has been provided.

Outputs

Output Distributions

One Customer Distribution column for each Attribute found in the Input Correlation Matrix.

Output Correlations

The Matrix Distributions node creates Normal (Gaussian) Customer Distributions with correlations according to the Input Correlation Matrix.

#4 Both Inputs

Shows the generated output when both the Input Attribute List (top-input) and the Input Correlation Matrix (bottom-input) is provided.

Inputs

Attributes

Input Attribute List.

Correlations

Input Correlation Matrix.

Node

Configuration

Select the name of the Attribute found in the Input Attribute List.

Outputs

Output Distributions

Output Customer Distributions.

Output Correlations

The correlations between the Output Customer Distributions.

Update #1 – Default Correlation

Default Correlation: The user can set the default Correlation for all generated Distributions from the Matrix Distributions Configuration Dialog. This default Correlation will be applied to all Customer Distributions found in the Input Attribute List and the Input Correlation Matrix. The default Correlation will be overridden by the row-column Correlation values found in the Input Correlation Matrix as long as both the [row][column] and [column][row] value are not missing. Uncorrelated Distributions will have a Correlation = 0.0, while perfect Correlation = 1.0.

 

Configuration Dialog

Correlation Matrix

Correlation Chart