Regression Model Nugget

A Regression model nugget represents the linear equation estimated by a Regression node. It contains all of the information captured by the linear regression model, as well as information about the model structure and performance.

When you run a stream containing a Regression model nugget, the node adds a new field containing the model's prediction for the output field. The name of the new field is derived from the name of the output field being predicted, prefixed with $E-. For example, for an output field named profit, the new field would be named $E-profit.

Generating a Filter node. The Generate menu enables you to create a new Filter node to pass input fields based on the results of the model. This is most useful with models built using one of the field selection methods. See the topic Regression Node Model Options for more information. For general information on using the model browser, see Browsing model nuggets .

Predictor Importance

Optionally, a chart that indicates the relative importance of each predictor in estimating the model may also be displayed on the Model tab. Typically you will want to focus your modeling efforts on the predictors that matter most and consider dropping or ignoring those that matter least. Note this chart is only available if Calculate predictor importance is selected on the Analyze tab before generating the model. See the topic Predictor Importance for more information.

Evaluating the Model

As with other generated models, you can use an Analysis node to evaluate the model results. You can also use a Plot node to display predicted values versus actual values, which can help you identify the records that are most difficult for the model to classify accurately and to identify systematic errors in the model.

You can also assess the linear regression 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 linear regression. See the topic Regression Model Nugget Advanced Output for more information.