Objectives (linear models)
What is your main objective? Select the appropriate objective.
- Create a standard model. The method builds a single model to predict the target using the predictors. Generally speaking, standard models are easier to interpret and can be faster to score than boosted, bagged, or large dataset ensembles.
- Enhance model accuracy (boosting). The method builds an ensemble model
using boosting, which generates a sequence of models to obtain more accurate predictions. Ensembles
can take longer to build and to score than a standard model.
Boosting produces a succession of "component models", each of which is built on the entire dataset. Prior to building each successive component model, the records are weighted based on the previous component model's residuals. Cases with large residuals are given relatively higher analysis weights so that the next component model will focus on predicting these records well. Together these component models form an ensemble model. The ensemble model scores new records using a combining rule; the available rules depend upon the measurement level of the target.
- Enhance model stability (bagging). The method builds an ensemble model
using bagging (bootstrap aggregating), which generates multiple models to obtain more reliable
predictions. Ensembles can take longer to build and to score than a standard model.
Bootstrap aggregation (bagging) produces replicates of the training dataset by sampling with replacement from the original dataset. This creates bootstrap samples of equal size to the original dataset. Then a "component model" is built on each replicate. Together these component models form an ensemble model. The ensemble model scores new records using a combining rule; the available rules depend upon the measurement level of the target.
- Create a model for very large datasets (requires IBM® SPSS® Statistics Server). The method builds an ensemble model by splitting the dataset into separate data blocks. Choose this option if your dataset is too large to build any of the models above, or for incremental model building. This option can take less time to build, but can take longer to score than a standard model. This option requires IBM SPSS Statistics Server connectivity.
See Ensembles (linear models) for settings related to boosting, bagging, and very large datasetsboosting and bagging.