Spring 2012
STAT 512
HW 5
1
Homework 5 (25.5 pts.)
due Feb. 24
A reminder – Please do not hand in any unlabeled or unedited SAS output. Include in your writeup only
those results that are necessary to present a complete solution (what you want the grader to grade). In
particular, questions must be answered in order (including graphs), and all graphs must be fully labeled
(main title should include the question number, and all axes should be labeled). Don’t forget to put all
necessary information (see course policies) on the first page. Include the SAS input for all questions at
the very end of your homework; this could be important even though it won’t be graded. You will often
be asked to continue problems on successive homework assignments so save all your SAS code.
The following problems are a continuation of Problem 3 in HW 4. Consider the data set from
Problem 6.18, p. 251 about “Commercial Properties” (CH06PR18.DAT) which describes a data
set (n = 81) used to evaluate the relation because rental rates (Y) and the age of the unit (X
1
),
operating expenses and taxes (X
2
), vacancy rates (X
3
), total square footage (X
4
).
SAS Code:
data
HW5P1;
infile
'I:\My Documents\Stat
512\CH06PR18.DAT'
;
input
rental age expense vacancy area;
run
;
proc
print
data
=HW5P1;
run
;
*a e;
Title1
'Rental Rates'
;
proc
reg
data
=HW5P1;
model
rental = age expense vacancy
area/
clb
;
output
out
=rentalout
r
= resid
p
=pred;
run
;
*c;
proc
corr
data
=HW5P1;
var
age expense vacancy area;
with
rental;
run
;
*d;
proc
corr
data
=HW5P1;
var
age expense vacancy area;
run
;
*f;
title2
'Residuals vs. predicted values'
;
symbol1
v
= circle;
proc
gplot
data
=rentalout;
plot
resid * pred/
vref
=
0
;
run
;
title2
'Residuals vs. age'
;
symbol1
v
= circle;
proc
gplot
data
=rentalout;
plot
resid * age/
vref
=
0
;
run
;
title2
'Residuals vs. expense'
;
proc
gplot
data
=rentalout;
plot
resid * expense/
vref
=
0
;
run
;
title2
'Residuals vs. vacancy'
;
proc
gplot
data
=rentalout;
plot
resid * vacancy/
vref
=
0
;
run
;
title2
'Residuals vs. area'
;
proc
gplot
data
=rentalout;
plot
resid * area/
vref
=
0
;
run
;
*g;
title2
'Normalality of Residuals'
;
proc
univariate
data
=rentalout
noprint
;
var
resid;
histogram
resid /
normal
kernel
(
L
=
2
);
qqplot
resid /
normal
(
L
=
1
mu
=est
sigma
=est);
run
;
*h;
data
a2;
age =
15
; expense =
9.2
; vacancy =
0.03
;
area=
90000
;
data
predict;
set
HW5P1 a2;
proc
reg
data
=predict;
model
rental = age expense vacancy
area/
cli
;
id
age expense vacancy area;
run
;
quit
;
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentSpring 2012
STAT 512
HW 5
2
1. (12 pts.) This problem relates to material described in Chapter 6 about multiple
regression.
(a) Run the multiple linear regression with age, operating expenses, vacancy rates
and total square footage as the explanatory variables and rental rate as the
response variable. From the data from the last problem set, finish the summary of
the regression results by giving the values of R
2
and the adjusted R
2
.
This was done in Homework 4:
This is the end of the preview.
Sign up
to
access the rest of the document.
 Fall '08
 Staff
 Regression Analysis, rental rate

Click to edit the document details