Statistics for Scientists & Engineers:
Exploratory Data Analysis
| Instructor: |
James J. Filliben |
| Textbook: |
None
|
| Format: |
5 sessions
2 to 3 sessions/week
3 hours/session
|
|
| Prerequisites: |
None |
| Registration Fee: |
None |
Exploratory Data Analysis (EDA) is an approach/philosophy
for data analysis which employs a variety of graphical
techniques to
- maximize insight into a data set,
- uncover underlying structure,
- detect outliers/anomalies,
- test underlying assumptions; and
- develop parsimoniuous models.
The EDA approach is contrasted with the classical
approach in both goal and methodology--this will be
discussed in detail. EDA methods to be discussed include
histograms, lag plots, probability plots, 4-plots, PPCC
plots, box plots, block plots, GANOVA plots, Youden plots,
interaction plots, residual plots, and general multiplots.
All graphical methods are illustrated through extensive
"real-life" data examples using the DATAPLOT
graphics/statistics language.
Check the SED Calendar for the
current schedule of upcoming courses.
If the course is not currently scheduled in the SED Calendar, please contact
Jim Filliben
for more information or to register scheduling requests.
The data sets used in this course
can be downloaded.
Date created: 6/5/2001
Last updated: 6/22/2001
Please email comments on this WWW page to
sedwww@cam.nist.gov.
|
