Icon

Most Important Regression Factors

Two workflows. First ranks by coefficents. Second ranks by either R^2 or Adjusted R^2. Thanks to @paolotamag for the R2 Adjusted metanode.

Adjusted R-squared is a modified version of R-squared that hasbeen adjusted for the number of predictors in the model. Theadjusted R-squared increases when the new term improves themodel more than would be expected by chance. It decreaseswhen a predictor improves the model by less than expected.Typically, the adjusted R-squared is positive, not negative. It isalways lower than the R-squared.Adding more independent variables or predictors to a regressionmodel tends to increase the R-squared value, which temptsmakers of the model to add even more variables. This is calledoverfitting and can return an unwarranted high R-squared value.Adjusted R-squared is used to determine how reliable thecorrelation is and how much it is determined by the addition ofindependent variables. Read WineQuality DataNode 2Node 3SelectCorrelation Filter70/3070/30Node 11Node 12Node 13RemoveInrerceptabs(coeff.)Node 23See NoteLoop ThroughFactorsNode 43Node 4770/30Node 49Node 55Add Factor NameNode 58Right Click toOpen Views CSV Reader Normalizer Linear Correlation Correlation Filter Partitioning Partitioning Linear RegressionLearner RegressionPredictor Numeric Scorer Row Filter Math Formula Sorter R2 Adjusted Column ListLoop Start Loop End Column Filter Linear RegressionLearner RegressionPredictor Sorter Variable toTable Row Column Appender Coefficient &R^2 Charts Adjusted R-squared is a modified version of R-squared that hasbeen adjusted for the number of predictors in the model. Theadjusted R-squared increases when the new term improves themodel more than would be expected by chance. It decreaseswhen a predictor improves the model by less than expected.Typically, the adjusted R-squared is positive, not negative. It isalways lower than the R-squared.Adding more independent variables or predictors to a regressionmodel tends to increase the R-squared value, which temptsmakers of the model to add even more variables. This is calledoverfitting and can return an unwarranted high R-squared value.Adjusted R-squared is used to determine how reliable thecorrelation is and how much it is determined by the addition ofindependent variables. Read WineQuality DataNode 2Node 3SelectCorrelation Filter70/3070/30Node 11Node 12Node 13RemoveInrerceptabs(coeff.)Node 23See NoteLoop ThroughFactorsNode 43Node 4770/30Node 49Node 55Add Factor NameNode 58Right Click toOpen ViewsCSV Reader Normalizer Linear Correlation Correlation Filter Partitioning Partitioning Linear RegressionLearner RegressionPredictor Numeric Scorer Row Filter Math Formula Sorter R2 Adjusted Column ListLoop Start Loop End Column Filter Linear RegressionLearner RegressionPredictor Sorter Variable toTable Row Column Appender Coefficient &R^2 Charts

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: genef
Created at: 2023-05-24
On NodePit since: 2023-05-28
Last update: 2024-12-21
Created with KNIME version: v4.7.3
Tags: @paolotamag

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!