Welcome to CSI 991

Seminar in Computational Statistics

This is a seminar series in the Computational Statistics Program in the CSI PhD program.

It is a working seminar in various topics in statistics. Each semester and during the summer, we choose some topic of interest and meet once a week (usually on Friday afternoon an hour or so before happy hour) to discuss the topic.

Sometimes we work through a recent monograph on the chosen topic, and sometimes we experiment with one or more computer packages. Sometimes we do small Monte Carlo studies, which participants may follow up with larger MC studies. We usually have a laptop with a projector so everybody can see the programs and the output.

The idea is to have some fun and to learn something.

Students can register for the seminar for 1 hour of credit. Students who register will be required to make presentations. Grades are assigned based on these presentations and general participation in the seminar. Many of the attendees, however, are just faculty or students who want to learn something and who don't need the credit. On the other hand, students who want to do a significant amount of work in the topic of the seminar may be allowed to register for 3 hours of credit in CSI 779, or STAT 789, or a variable number of credits in CSI 796. Students registered for these courses meet with the seminar group and also individually with the instructor to work on an assigned research project.

Anyone is welcome to attend.


Spring, 2015: CSI 991, Section 001: Recent articles in computational learning and computational statistics.
Time and place: Fridays from 3:00 to 5:00 in Innovation Hall 338.


Sometimes we make web pages and sometimes we don't and some are more complete than others.
For the past few semesters, we have run the seminar as a journal club.
Fall, 2014, Spring, 2014 Fall, 2013.


Spring, 2013: Evidence in statistical inference.
Fall, 2012: Recent articles in computational learning and computational statistics.
Summer, 2012: Recent articles in computational learning and computational statistics.
Spring, 2012: Comparison of Bayesian and frequentist methods in statistics
Fall, 2011: Bayesian Methods in Statistics
Spring, Summer, 2011: Latent Variable Models
Fall, 2010: Bayesian Computations in R
Spring, 2010: Statistical Learning
Fall, 2009: Statistical Learning
Summer, 2009: Statistical Learning
Spring, 2009: Statistical Computing in R
Fall, 2008: Statistical Computing in R
Spring, 2008: Statistical Learning.
Fall, 2007: Statistical Learning.
Spring, 2007: Pattern classification
Fall, 2006: Linear Models in R.
Summer, 2006: R Graphics.
Spring, 2006: Topics in the Exploration of Data.
Fall, 2005: Topics in the Exploration of Data.
Spring/Summer, 2005: Topics in the Exploration of Data.
Fall, 2004: Topics in the Exploration of Data.
Spring/Summer, 2004: Data Mining.
Fall, 2003: Data Mining.
Summer, 2003: Logistic Regression.
Spring, 2003: New (and Old) Ways of Looking at Regression (Continuation of topics from Fall, 2002.)
Fall 02: New (and Old) Ways of Looking at Regression
Spring 02: Statistical Learning
Fall 01: Classification and Regression Trees; CART and MARS
Summer 01: More on Robust Statistical Methods
Spring 01: Robust Statistical Methods
Spring 96: Alternatives to Least Squares II
Fall 95: Alternatives to Least Squares
Spring 95: Fitting Generalized Additive Models

Send suggestions to jgentle@gmu.edu or to csutton@gmu.edu.