W4315 Linear Regression Models

Course Description

This course will teach the theory and practice of regression analysis, for both simple and multiple regression, including testing, estimation, and confidence procedures, modeling, regression diagnostics and plots, polynomial regression, colinearity and confounding, model selection, geometry of least squares. Extensive use of the computer to analyze data.

Lectures

• Section 1: MW 8:40am-9:55am 428 Pupin Laboratories

• Section 2: MW 6:10pm-7:25pm 329 Pupin Laboratories

Textbook

• (Required) Applied Linear Regression Models 4th Ed., by Kutner, Nachtsheim, and Neter. McGraw-Hill, 2004. Available on Amazon.com NOTE, DO NOT PURCHASE THE 5TH EDITION Applied Linear Statistical Models

• (Recommended): Linear models with R, by Julian J. Faraway. Boca Raton :Chapman & Hall/CRC, 2005.

Prerequisite

• Calculus

• Probability and statistics

• Linear algebra

Office hour

• Yang Feng: Monday 10am-noon, SSW 1012

• Ran He (Email: ran@stat.columbia.edu)

1. Monday 7:30 - 9:30 pm, SSW 1025

2. Tuesday 2:30-4:30pm Stat Dept. Lounge, 10th floor of SSW

• Yuting Ma (Email: ym2396@columbia.edu)

1. Tuesday 12:30 - 2:30pm, Stat Dept. Lounge, 10th floor of SSW

2. Friday 9:30 - 11:30 am, Stat Dept. Lounge, 10th floor of SSW

Grading

• Homework (20%), due 6pm every Wednesday.

• Midterm (30%), Mar. 13, Wednesday, in class. Section 1 Statistics Section 2 Statistics

• Final Exam (50%)

• Section 1: May 15, 9:00am-Noon 428 Pupin Laboratories

• Section 2: May 13, 7:10pm-10:00pm 329 Pupin Laboratories

Homework

Weekly. Contains a mix of paper and computer problems. The assignments will be posted on the course page every Wednesday and due 6PM the next Wednesday unless stated otherwise. When submitting the homework, please place your homework in the W4315(1)-Inbox / W4315(2)-Inbox (depending on your corresponding section) for the course at SSW 904, 1255 Amsterdam Ave. PLEASE DO NOT HAND IN THE HOMEWORKS IN CLASS. Graded homework will be placed in W4315(1)-OutBox/ W4315(2)-Outbox. Late homework will receive a grade of zero. To compensate, the lowest homework score will be dropped. To receive credit on homework, you must show all work neatly, clearly label each problem, and staple the entire assignment together in correct order with your name printed. The homework will be graded, and each homework problem carries equal weight. You are allowed to work with other students on the homework problems, however, verbatim copying of homework is absolutely forbidden. Therefore each student must ultimately produce his or her own homework to be handed in and graded.

Any questions regarding homework grading, please contact

• Di Cheng (dc2884@columbia.edu) for Odd Numbered Homework (HW 1, 3, 5, 7, 9)

• Yongjoon Kim (yk2494@columbia.edu) for Even Numbered Homework (HW 2,4,6,8,10)

Exams

There will be one in-class midterm exam, and a final exam. Both of them are closed book and closed notes. Both exams are required and there will be no make-up exams. Missed exams will receive a grade of zero. In each exam, you may take one letter size paper (you can write on both sides) as a cheat sheet. Non-graphing calculators may be used during the exams. Any questions regarding exam grades should be submitted in writing (a paper briefly stating where is the grading error with your signature and date, along with your exam) within one week of distribution of the graded exam. You must hand in the request to the instructor before or after class. Notice that any request will lead to the reevaluation of the whole exam and your total score may increase or decrease. While all numeric grades will be the raw grades, the final letter grades will be curved.