Advanced Data Analysis I

CMU 36-401 (Fall 2000)

Jump to: Schedule/HW     Data     Handouts     Splus    

Professor Howard Seltman
232H Baker Hall
268-3938
hseltman@stat.cmu.edu
Office hours: 1-2PM Week of 12/11 M,T,W,F
Class Schedule Tuesday & Thursday, 3-4:20 p.m.
DH 2105
Syllabus & Policies html form
postscript form
Teaching Assistant B. Ricky Rambharat
A60-A Baker Hall
268-6289
ricky@stat.cmu.edu
office hours: Wed 3:30-4:30, Wean 5202


Schedule and Homework

Date Subject Material Reading Homework
Tu 8/29 Course introduction, Examples    
Th 8/31 Introduction to Splus (Meet in Wean 5202) MASS pp. 1-68: skim headings Homework 1 Due Th 9/7     Solutions
Tu 9/5 Exploratory data analysis RwG pp. 1-23  
Th 9/7 - Th 9/12 Simple linear regression RwG pp. 29-59 Homework 2 Due Th 9/14     Solutions
Th 9/14 Simple linear regression RwG pp. 29-59 Homework 3 Due Th 9/21     Solutions     Excellent Report
Tu 9/19 Multiple regression RwG pp. 65-84  
Th 9/21 Splus Skills (Meet in Wean 5202)   Homework 4 Due Th 9/28     Solutions
Tu 9/26 - Th 9/28 Multiple regression RwG pp. 65-84 Homework 5 Due Th 10/5     Solutions  (Excellent Pr. 4
Tu 10/3 Interaction / Dummy variables RwG pp. 84-92  
Th 10/5 Example / Review (skip RwG pp. 92-101) (No Homework Due Th 10/12)
Tu 10/10 MIDTERM EXAM    
Th 10/12, Tu 10/17 Regression criticism RwG pp. 109-125 Homework 6 Due Th 10/19   (pdf version)     Solutions
Th 10/19, Th 10/24 Outliers: influence and potential / fitting curves RwG pp. 125-137, 145-163 Homework 7 Due Th 10/26     Solutions
Th 10/26 Nonlinear regression RwG pp. 163-174 Homework 8 Due Th 11/2     Solutions
Tu 10/31, Th 11/2 Journal Articles Peptide Drug Delivery
Trickle Down
Homework 9 Due Th 11/9
Tu 11/7, Th 11/9 Project description presentations Group Project Selection Form Homework 10 Due Th 11/16
Tu 11/14, Th 11/16 Robust regression RwG pp. 183-212 Homework 11 Due Th 11/30
Tu 11/21 Random effects and Mixed models Tutorial     Paper (hw11 due 11/30)
Th 11/23 Happy Thanksgiving! Turkey   Turkey Leg  
Tu 11/28 Th 11/30 Logi(s)t(ic) regression RwG pp. 217-242 Homework 12 Due Th 12/7
1st e-mail project report due 2 PM Th 11/30
Tu 12/5, Th 12/7 Monte Carlo and Bootstrap methods RwG pp. 303-326 2nd e-mail project report due 2 PM Th 12/7
Tu 12/12 Review for FINAL EXAM   Group project written report due 3 PM
Wed. 12/20 1-4PM FINAL EXAM Baker Hall A53  
(*) Project progress meetings with instructor 11/27/00 to 12/6/00.


Data Downloads

Link Description More info
concord1.dat Concord household water use concord1.txt
widgets.dat Widget data for HW 1, problem 2  
seapart.dat Seabird data (partial); for HW 1, problem 3 See RwG p. 61
cane.dat Sugar cane data for HW1, problem cane.txt
cyclist.html Cyclist data from 9/7 lecture  
homedat.html Home resale data from 9/7 lecture  
fuel.dat Consumer reports data for HW2, problems 1 and 2 Columns: Weight Disp Mileage Fuel Type
lead.dat Lead toxicity data for HW2, problem 3  
televisions.dat Life expectancy data for HW3 televisions.txt
wages.dat CPS wage data for 9/19 lecture and HW4 wages.txt
cheese.dat Cheese data for 9/21 S-plus skills class cheese.txt
ozone.dat Ozone data for 10/5 review ozone.txt
salamander.dat Salamander data for hw6 See RwG p. 104
cassava.dat Cassava data for hw7 See RwG p. 137
deforest.dat Deforestation data for hw8 #1 See RwG p. 175
solrad.dat Solar radiation data for hw8 #3 See RwG p. 177
eggs.dat Egg data for hw8 #5 See RwG p. 180
kob.dat Kob data for hw11 #1 See RwG p. 213
titanic.dat Titanic data for hw12 Columns: Class, Adult, Male, Survive


Handouts and In-Class Exercise

Link Description More info
0829.ps 8/29 Notes and breakout Article
Unix.ps Unix Summary  
Sintro.ps 8/31 Notes: Introduction to S-plus Exercise solutions: Sintro.q
0905.ps 9/5 Notes: Graphs for univariate distributions (RwG Ch. 1)  
0907.ps 9/7 Notes: Simple Regression I (RwG Ch. 2)  
0912.ps 9/12 Notes: Simple Regression II (RwG Ch. 2)  
0914.ps 9/14 Notes: Simple Regression III (RwG Ch. 2)  
0919.ps 9/19 Notes: Multiple Regression I (RwG Ch. 3)  
sskills1.q 9/21 S-plus skills command file #1  
sskills2.q 9/21 S-plus skills command file #2  
Latex.ps Intro to using Latex for reports template.tex
0926.ps 9/26 Notes: Multiple Regression II (RwG Ch. 3)  
0928.ps 9/28 Notes: Multiple Regression III (RwG Ch. 3)  
1003.ps 10/3 Notes: Interaction and Dummy Variables (RwG Ch. 3)  
1005.ps 10/5 Notes: Example and Review (RwG Ch. 1-3)  
1012.ps 10/12 Notes: Assumptions I (RwG Ch. 4)  
1017.ps 10/17 Notes: Assumptions II (RwG Ch. 4)  
1019.ps 10/19 Notes: Influence and potential (RwG Ch. 4)  
1024.ps 10/24 Notes: Influence / curvilinear regression (RwG Ch. 4/5)  
1026.ps 10/26 Notes: Nonlinear regression (RwG Ch. 5)  
1114.ps 11/14 Notes: Robust Regression I (RwG Ch. 6)  
1116.ps 11/16 Notes: Robust Regression II (RwG Ch. 6)  
1121.ps 11/21 Notes: Mixed Models (Not in RwG)  
1128.ps 11/28 Notes: Logisitic Regression I (RwG Ch. 7)  
1130.ps 11/30 Notes: Logisitic Regression II (RwG Ch. 7)
1205.ps 12/05 Notes: Computer Intensive Methods I (RwG Appendix 2)  
1207.ps 12/07 Notes: Computer Intensive Methods II (RwG Appendix 2)  
1212.ps 12/12 Notes: Review  


Splus Downloads

Link Description More info
cheatsheet Splus usage summary  
contents.html Splus on-line tutorial  
SplusTips.html Splus Tips and Links  
symplot.q function: symmetry plot  
ps.q function: copy a plot to a .ps file and optionally print it
Sintro.q commands: Sintro.q Solutions to Splus Intro exercises
ndhist.q function: histogram plus normal density  
rwgbox.q function: boxplots in RwG style  
qplot.q function: quantile plot  
qqn.q function: quantile-normal plot  
scatbox.q function: scattergram + marginal boxplots  
collin.q function: check collinearity for a set of X's R^2 for predicted each on all others
sum.step.q function: exhaustive model selection by AIC, BIC or adj. R^2  
mypairs.q function: pairs plot with boxplots for few categories  
DW.q function: Durbin-Watson test  
spruce.q sample code: for hw6 (dummy/interaction plotting)  
influence.q function: influence measures  
partreg.q function: partial regression influence plots  
CookPlot.q function: residual vs. fit plot with Cook Distance  
medplot.q function: Exploratory band regression  


UP to Howard's Home Page