turn on suggestions

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
- /
- SAS/GRAPH and ODS Graphics
- /
- KM Plot line thickness

Topic Options

- 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

02-05-2014 12:29 AM

Hi there,

I am using SAS 9.3 ( recently moved from SPSS to SAS.).

I am running KM survival curve and my questions is slightly "cosmetic" .

I want the KM survival lines **thicker** (double thicker than default) and different colors than default. Below is the code.

In output, I am having KM survival plot in perfect shape but not with the thickness and line colors I wanted.

Your help is much appreciated.

E.K.

***KM test;

goptions lfactor=5 vsize=5 hsize=7 ftext=swissb noprompt;

symbol1 c=black v=point h=1 i=steplj l=1;

symbol2 c=blue v=point h=1 i=steplj l=1;

symbol3 c=red v=point h=1 i=steplj l=1;

symbol4 c=orange v=point h=1 i=steplj l=1;

axis1 label=('Years')

order=(0 to 7 by 1)

minor=none;

axis2 label=(a=90 'Proportion Free from Death')

order=(0 to 1 by 0.2)

minor=none;

Title 'XYZ';

proc lifetest data=dt plots=survival(atrisk (maxlen=13 outside(0.15)));

time KM_years * censor(0);

strata abc /order=internal;

run;

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

Posted in reply to statkoc

02-05-2014 07:22 AM

Hi.

One way would be to use the ODS table to output the survival data in a SAS table, then use PROC (S)GPLOT to plot it.

If you run the ODS TRACE ON; command before your proc lifetest. in the log file you will get the SAS tables associated with this PROC.

Then run ODS TRACE OFF to stop getting the information in the log;

Anca.

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

Posted in reply to AncaTilea

02-05-2014 09:11 AM

Here is a blog article on this. In the STEPPLOT, change the thickness of the line by using the following option: STEPPLOT x= y= / LINEATTRS=(THICKNESS=2).

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

Posted in reply to statkoc

02-05-2014 08:30 AM

Statkoc,

You are working with ods graphics off and trying to apply xchanges to classical sas graphs

http://www.nesug.org/Proceedings/nesug06/po/po08.pdf

but there is another manner of working i shall try to explain here

(working by default for 9.3.2 as ods graphics on; is active)

Lets take few similar code ods graphics must be open !!!!!

first line open the sge possibility

second line rtestrict the work to only the graph plot (without any sas graph instructions):

ods html sge=on;

ods select lifetest.survivalplot;

**proc** **lifetest** data=muco.survie_cohorte_sexe graphics plots=(s) cs=none width=**1** ;

time duree_vie*censure(**0**) ;

strata cohorte_sexe ;

survival out=muco.sortie_survie_cohortes_sexe_mkm ;

title 'Analyse de survie par cohorte de sexe 1' ;

title2 'Methode Kaplan-Meier' ;

**run** ;

ods select all;

you have some results 2 thin lines and colored in blue and red

Now open the sge **sas graphic editor **

and you can change the style from statistical to analysis

and the thickness of your graphic and a lot of others details on the two types of line: scatter and survival

with a right clic on a line

and saving it in png under a new name

You have a lot of different examples on survival code on pages 124-137 in my text

http://www.ined.fr/brochure_wielki/sasvolume2

BUT it is written in french

Some texts in english are indicated here :

**about the SGE**

Sanjay Matange ODS GRAPHICS EDITOR http://www.nesug.org/Proceedings/nesug08/po/po24.pdf

**about survival** see here a very good resume of ways

http://www.pharmasug.org/proceedings/2012/DG/PharmaSUG-2012-DG05.pdf

Finally you have also the **ods graphical designer** as clic click solution see visual pages 138 to 141

the learning process could take time as you are starting in sas coming from spss.

a lot of sas users are not knowing yet what is explained here.

Good luck

HTH

Andre