-
Nested Test Tool
Nested Test Tool
One Tool Example
Nested Test has a One Tool Example. Visit Sample Workflows to learn how to access this and many other examples directly in Alteryx Designer.
The Nested Test tool examines whether two models, one of which contains a subset of the variables contained in the other, are statistically equivalent in terms of their predictive capability. The larger of the 2 models is referred to as the "full" model, while the model that contains a subset of the variables is referred to as the "reduced" model. In the statistics literature, the "reduced" model is often called the "null" model, while the "full" model is called the "alternative" model.
In the case of 2 linear regression models, the test is conducted using an F-test*. For 2 GLM family models, including count regression, gamma regression, and logistic regression, the test is conducted using a likelihood ratio test**.
This tool uses the R tool. Go to Options > Download Predictive Tools and sign in to the Alteryx Downloads and Licenses portal to install R and the packages used by the R tool. Visit Download and Use Predictive Tools.
Connect Inputs
The output stream from either a Linear Regression tool or a GLM family tool (Count Regression tool***, Gamma Regression tool, or Logistic Regression tool) used to create the "full" model, the one with a greater number of variables. The stream can be entered as any of the three inputs.
The output stream from either a Linear Regression tool or a GLM family tool (Count Regression tool***, Gamma Regression tool, or Logistic Regression tool) used to create the "reduced" model, the one with a subset of the variables included in the "full" model. The stream can be entered as any of the three inputs.
The Alteryx data stream that was used to create both the "full" and "reduced" models. The stream can be entered into either side of the tool.
Configure the Tool
This tool has no configuration options.
*https://en.wikipedia.org/wiki/F-test **https://en.wikipedia.org/wiki/Likelihood-ratio_test ***Quasi-poissonmodel not supported.