## Statistics 410: Statistical Computing and Linear Models

Time & Location: TTh 9:25AM-10:40AM HB21

Required Text: Applied Linear Statistical Models (4th ed), by Neter, Kutner, Nachtsheim, and Wasserman.

Supplementary Text: Modern Applied Statistics with S-Plus (3rd ed), by Venables and Ripley.

Instructor: Keith A. Baggerly

Office: 2100 DH

phone: 5282 (longphone: +1-713-348-5282)

email: kabagg@stat.rice.edu

50% Homework: assigned approximately weekly on Thursday, due following Thursday. There will be a 20% deduction for each late day after the deadline. You are allowed to allowed to collaborate with your classmates on the homework problems. Homework Problems will require the use of S-Plus, MATLAB or a similar computational engine.

50% Exams: there will be two exams, one roughly at midterm, and the other at the end of the semester. The exams will be take-home, and will have a computer component to them. I reserve the right to turn the second exam into a report (I will discuss this in class).

Data & Code:

I will use this location to post large data sets that we will analyze (to remove the need for typing everything in). I will also occasionally post snippets of code here (with comments) to illustrate aspects of S-plus or Matlab.

Homework:

• HWK 1 (due 1/27): Chapter 1: 27, 41, 42a-c. Chapter 2: 27, 28, 29. Dot products (from class Thursday).

• HWK 2 (due 2/8): Graphing a regression function in Splus. You should write an Splus function that takes two vectors, x and y, and returns: a 2 by 2 matrix of plots, the upper left should be an x-y plot of the data with the regression line superimposed, the upper right should be the regression line with 90% confidence bands for the line and for an individual observation shown, the lower left should be a plot of residuals versus fitted y values, and the lower right should be a normal QQ plot of the residuals. The function should also return the numerical values of b0 and b1, the MSE, and t-values for b0 and b1. A data set to try this on is given here. I need the plot, the code, and the nuerical results.

• HWK 3 (due 2/22): The homework is described in the Data file here.

• HWK 4 (due 3/2): The homework is described in the Data file here.

• HWK 5 (due 3/21): The homework is described in the Data file here.

• Exam 1(Due 3/23): Description

• HWK 6 (due 4/11): The homework is described in the Data file here.

• HWK 7 (due 4/27): The homework is described in the in class handout. No data here this time!

• Exam 2 - The Final (Due end of finals) : Description

All graded homeworks will be in a box outside of 2100 Duncan Hall.