STAT 705 Spring 2009
------------------

Homework 4
----------

Do the following problems from the textbook:

19.1, 19.2, 19.3, 19.4, 19.5, 19.7, 19.13(b,c,d), 19.20(c,d), 19.21(b,c,d), 19.31(b), 
19.35(b,e), 19.48, 20.1, 20.2(a,b), 20.4, 23.10(c,d,f), 24.7(b,c,d,e).

NOTE: Problems 19.13, 19.20, 19.21, 19.31, 19.35, 20.2, 20.4, 23.10, 
and 24.6 require analysis using a computer package such as SAS or R.

HINT:  For 19.3, use equation (19.8a) on page 823.

NOTE: For 19.20(c,d), do the residual plot (vs. fitted values)
and the normal Q-Q plot of the residuals and comment on your findings.
Also, do the Brown-Forsythe test and the Shapiro-Wilk test on the
residuals and comment on your findings.

NOTE: For 19.35(e), use the Tukey procedure to investigate differences
in each pair of treatment means.

NOTE: For 20.2(b), don't worry about the Kimball inequality.

NOTE: For 23.10(c,d), the book mentions "full model" and "reduced model", 
but all I want you to do is fit one model and do the interaction-effects 
F-test (for part c) and main effects F-tests (for part d) based on that one 
model fitting.  You don't have to fit multiple models because the Type III SS 
section does that internally, automatically. 
(If you're interested in "full vs. reduced", details are on p. 955-959.)

The "Eye Contact Effect", "Programmer Requirements", 
"Coin-operated Terminals", "Adjunct Professors", 
and "Case Hardening" data sets are given on 
the course web page.

Please write your answers neatly and clearly!