Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- SAS Programming
- /
- General Programming
- /
- Partial F test and dummy variables

- RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

04-30-2016 05:03 PM - edited 05-02-2016 12:14 PM

data one;

input healthpoints age housetype;

datalines;

114 66 3

242 38 1

232 31 2

90 57 2

282 24 1

122 68 1

207 60 1

222 23 2

238 49 1

262 31 1

262 26 2

47 70 2

178 40 3

212 31 3

247 20 1

125 40 1

128 66 1

178 45 2

202 40 2

;

a)what is the point estimate of the age effect for each housing types .(am not sure what point estimate means in this scenario..do they mean parameter estimate? ....maybe it's like one dummy varaible per house type)

b)Is the point estimate sig. different between housing=1 and housing=2. ( use either partial f test or t test maybe)

c) propose a model to test all three housing to see if they are sig different from each other (yes again...similiear to b but testing all at same time, this is related to interaction modelling)

----------Below are codes that may be helpful in solving the problems

subsetting for categorical that has 3 categories

**data** a;

set one;

if x=**1** or x=**2**;

**run**;

subsetting for categorical that has 2 categories

**data** b;

set a;;

if x=**1** then dum=**1**;

else dum=**0**;

**run**;

intereaction term for one continous variable and one dummy variable

**data** c;

set b;

zdum=z*dum;

**run**;

two dummy variable for a cat. variable with 3 categories

**data** aa;

set one;;

if x=**1** then dum1=**1**;

else xdum1=**0**;

if x=**2** then dum2=**1**;

else dum2=**0**;

**run**;

intereaction for continues with two dummy variables

**data** bb;

set aa;

zdum1=z*dum1;

zdum2=z*dum2;

**run**;

sample of partial f test

**proc** **reg** data=d;

model y=x1 x2 x3;

test x2=x3=**0**;

**run**;

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to BruceAlmighty

05-02-2016 11:36 AM

Please provide example in the form of a data step if you want someone to test with your data. Since Excel has no actual data types then the choices others make to import your data into SAS may result in differing data types. Which means that their code may not run with your SAS dataset. Also some of our users are blocked from downloading Excel files by their security policies.

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to ballardw

05-02-2016 12:14 PM

Thanks for letting me know. I have edited my post.