¶ Description
Compute Linear Model and MANOVA.
¶ Parameters
¶ Parameters tab
Parameters:
- Input table partitioning (1st pin)
- Predictors numeric
- Predictors nominal
- Target variable
- Mod: family
- Mod: include intercept?
- Mod: scale predictors (standardize)
- Mod: random seed
- Mod: leaning sample proportion
- Mod: numerical missing values inputation
- Mod: text missing
- Mod: variable selection
- Mod: stepwise method
- Mod: cross validation
- Out_ number of outliers to dipslay on plots
- Out: plot only first n rows
- Out: model file: name (leave blank to use target name)
- Out: model file: include time stamp in name
- Out: output type
- Out: save model
- Out: export model to pmml
- Out: include prediction in first output
- Predictor variables from table
- A short descriptiontitle: R_GLM2
¶ Configuration tab
See dedicated page for more information.
¶ About
This node provides a combination of LM (Linear Model) and GLM (Generalized Linear Model) functionalities.
LM is used to fit linear models. It can be used to carry out regression, single stratum analysis of variance and analysis of covariance.
GLM can either compute a single model, or work as a “model factory” when the “partition by column” mode is selected.
In this example, we prepare data using a “Time Travel” transformation, and we have several time series (per location, brand, segment… you name it), and we want to have one predictive model for each series, in a single operation.
We simply sort the data using the partition variable, and properly set the GLM node:
