Naomi Altman
(Penn State Statistics Department)
Interpreting and Extending Principal Components Analysis
What | Meteo Colloquium UG Homepage GR |
---|---|
When |
Feb 18, 2015 03:30 PM
Feb 18, 2015 04:30 PM
Feb 18, 2015 from 03:30 pm to 04:30 pm |
Where | 112 Walker Building |
Contact Name | Martin Tingley |
Contact email | mpt14@psu.edu |
Contact Phone | 814-865-0479 |
Add event to calendar |
![]() ![]() |
Principal Components Analysis (PCA) is often used to summarize high dimensional data. The leading principal components are often used as descriptors of the principal modes of variation in the data, or interpreted as observable patterns. In this talk I will discuss the role of PCA in dimension reduction, the interpretation of the leading eigenvectors and the possibilities of extending PCA to more complex types of data.