STAT 789

Advanced Topics in Statistics: Computer-intensive Methods for Classification and Regression

Instructor: Clifton D. Sutton

Texts:

• The Elements of Statistical Learning: Data Mining, Inference, and Prediction by Hastie, Tibshirani, and Friedman
• Classification and Regression Trees by Breiman, Friedman, Olshen, and Stone

Prerequisite:

Description:

Approximate week-by-week content:

[1] Tu June 7:

introduction to prediction and modeling in classification and regression settings --- comparing model-based methods to locally-adaptive methods

[2] Th June 9:

linear methods for regression, and regression modeling strategies

[3] Tu June 14:

linear methods for classification (including linear discriminant analysis and logistic regression)

[4] Th June 16:

tree-based methods for classification and regression (CART)

[5] Tu June 21:

more on CART

[6] Th June 23:

basis expansions, splines (linear splines, cubic splines, and MARS), and regularization

[7] Tu June 28:

more on MARS

[8] Th June 30:

kernel methods (local regression, density estimation and mixture models for classification, naive Bayes classifiers)

[**] Tu July 5:

(No class due to 4th of July holiday break

[9] Th July 7:

model assessment and selection (cross-validation, bootstrap methods, and other ways of evaluating model complexity and prediction accuracy)

[10] Tu July 12:

perturb and combine methods and ensemble classifiers (bagging, boosting, arcing, and random forests)

[11] Th July 14:

forward stagewise additive modeling (TreeNet)

[12] Tu July 19:

neural networks, support vector machines, and prototype methods

[13] Th July 21:

presentation of regression projects, review for exam

[14] Tu July 26:

presentation of classification projects, review for exam

[**] Th July 28:

Final Exam (note: exam period is from 7:30 to 10:15 PM)

Grading:

• 10% for useful participation (which is not the same thing as talking)
• 40% for presentation (oral and written) of project (which will be a thorough analysis of a data set that you choose (but I agree on), using several of the methods emphasized in class)
• 50% for open book (and notes) final exam

Additonal Comments:

• Put STAT 789 in the subject line when you send me e-mail (due to spam, I sometimes delete messages without reading them, based on the subject line).
• Be sure to note that there is not a class meeting scheduled for July 5 (due to 4th of July holiday break). However, if any class meetings are canceled prior to July 5 (perhaps due to power outage), it could be that July 5 will be used to make up for the missed class.
• I can possibly make arrangements to meet with you outside of my scheduled hours; however, on Tuesdays and Thursdays I do not like to be bothered from 7:00 to 7:17 and on Fridays I'm tied up with other activities for most of the day.
• Please do not leave long messages on my voice-mail, and since I often don't get around to returning calls until the evening, you should state what time you plan to go to sleep. Always leave your phone number, speaking slowly, even though you might have given it to me previously. I find it better to communicate with people in person or via e-mail --- phone tag is frustrating and sometimes the GMU voice-mail system doesn't work the way it is supposed to.
• You are expected to familiarize yourself with the George Mason University honor code and abide by it. In particular, it will be considered to be a violation of the honor code if you give or receive aid on the final exam.
• You are expected to take the final exam during the designated time slot; Incompletes will not be granted except under very unusual circumstances.
• Please abide by the university policy that cell phone ringers be turned off while class is in session.
• Note that eating and drinking in electronic classrooms is not allowed --- please do not violate the rules. (We will have a 10 minute break during each lecture period.)
• Any class meetings canceled by the university due to snow, sleet, power outage, bombing, etc. will be made up if possible (at a time to be agreed upon by me and as many members of the class as is possible if the univeristy doesn't specify a particular make-up date). With regard to bad weather, I will plan to teach class if the university is open and not teach it if the university is closed. So instead of calling me to find out if I plan to have class, just find out if the university is open or closed.
• Caveat: The schedule and procedures described here for this course are subject to change (and it is the responsibility of students to attend all class meetings and keep themselves informed of any changes).