Hello,
Can someone help me answer these questions regarding standardization and transformation?
- Does standardization handle the problem of extreme observations?
- Can Standardization be done instead of transformation?
- Will it be right to do variable transformations and then variable standardization in the same analysis?
Thanks,
Standardization is usually done to provide better mathematical properties in some algorithms, or to put variables on the same scale if they were on very different scales unstandardized. Transformation changes the distribution of the data.
Standardization is usually done to provide better mathematical properties in some algorithms, or to put variables on the same scale if they were on very different scales unstandardized. Transformation changes the distribution of the data.
This post about standardization might be helpful: https://communities.sas.com/t5/SAS-Communities-Library/To-standardize-data-or-not-to-standardize-dat...
SAS Innovate 2025 is scheduled for May 6-9 in Orlando, FL. Sign up to be first to learn about the agenda and registration!
Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.
Find more tutorials on the SAS Users YouTube channel.