Regularized regression open-source extension procedure enhancements in SPSS Statistics

SPSS Statistics

Your hub for statistical analysis, data management, and data documentation. Connect, learn, and share with your peers!

View Only

Summary

The latest version of IBM SPSS Statistics introduces a host of new statistical procedures, usability improvements, and improved open-source extension integration.

In this webcast, you’ll learn about three new Python-based regularized linear regression commands for handling models with highly correlated predictors, or even singularities resulting from problems with more predictors than cases.

Key takeaways:

Learn about new commands introduced within SPSS Statistics 29 for regularized linear regression modeling, fitting models using ridge, LASSO, and elastic net estimation
See how models can be fitted using specific values of regularization parameters, produce trace plots of regression coefficients for specified sets of regularization parameters, and select values of regularization parameters using cross-validation.
Simplify specification of analyses using command syntax or via graphical interfaces, so that analyses and output work as they do for native procedures
Key Speakers

David Nichols - Lead Statistician, SPSS Statistics

David holds a Ph.D. in Research Methodology and Quantitative Psychology from the University of Chicago. He leads statistical planning and design for IBM SPSS Statistics. He previously was the Lead Statistician for Watson Machine Learning Visualization.