GHUCCTS

Event Calendar

Share This

MHRI-GHUCCTS Monthly Statistical Seminar Series: Classification and Regression Trees

Date Fri, Mar 21
Time 12: 00 PM - 1: 00 PM
Location Zoom


This month's MHRI-GHUCCTS Monthly Statistical Seminar Series will be held Friday, March 21, 2025, from 12 PM - 1 PM. This series features Dr. Paul Kolm, Associate Director in the Center for Biostatistics, Informatics, & Data Science (CBIDS) at MedStar Health Research Institute.

The topic of this month's seminar will be tree-based analysis assesses relationships among variables by dividing the variables into subcategories of categorical variables or “bins” for continuous variables that optimizes prediction of the outcome variable (can be categorical or continuous). This is in contrast to developing mathematical prediction models. Some advantages of tree-based models are eliminating the need for variable transformation (no assumptions about variable distribution) and allowing for missingness to be included as a category itself.  

  

Upcoming Events
This free Spring Training Series will cover topics including, Foundations of Data Science, Introduction to Python, Basic Statistical Concepts, Data Exploration & Visualization, Experimentation in Data Science, and introduction to algorithmic techniques in Machine Learning.
This free Spring Training Series will cover topics including, Foundations of Data Science, Introduction to Python, Basic Statistical Concepts, Data Exploration & Visualization, Experimentation in Data Science, and introduction to algorithmic techniques in Machine Learning.
Mar 20
This symposium explores how experts from different fields—ethics, data science, informatics, and epidemiology—work together to make sure health data are used responsibly. We will discuss real-world challenges like protecting people’s privacy, getting consent before using personal data, and ensuring that health decisions are fair for all communities, especially those who have been—and continue to be—underrepresented, underprivileged, or underserved (the "triple Us").
Close