PCA/Factor Model Nugget

A PCA/Factor model nugget represents the factor analysis and principal component analysis (PCA) model created by a PCA/Factor node. They contain all of the information captured by the trained model, as well as information about the model's performance and characteristics.

When you run a stream containing a factor equation model, the node adds a new field for each factor or component in the model. The new field names are derived from the model name, prefixed by $F- and suffixed by -n, where n is the number of the factor or component. For example, if your model is named Factor and contains three factors, the new fields would be named $F-Factor-1, $F-Factor-2, and $F-Factor-3.

To get a better sense of what the factor model has encoded, you can do some more downstream analysis. A useful way to view the result of the factor model is to view the correlations between factors and input fields using a Statistics node. This shows you which input fields load heavily on which factors and can help you discover if your factors have any underlying meaning or interpretation.

You can also assess the factor model by using the information available in the advanced output. To view the advanced output, click the Advanced tab of the model nugget browser. The advanced output contains a lot of detailed information and is meant for users with extensive knowledge of factor analysis or PCA. See the topic PCA/Factor Model Nugget Advanced Output for more information.