topic Re: Proc Rank in SAS Procedures

Proc Rank

newguy1580 — Sat, 15 Mar 2014 23:50:08 GMT

Hi everyone,

I have a question regarding proc rank. I created quartiles in sas using proc rank and then ran phreg afterwards for my model. my output however has 2 parameter estimates instead of 3 (I set the lowest quartile as the reference) Can anyone explain to me why that I have only 2 parameter estimates? Thanks!

Re: Proc Rank

newguy1580 — Sun, 16 Mar 2014 02:10:03 GMT

Anyone?

Re: Proc Rank

Reeza — Sun, 16 Mar 2014 03:01:00 GMT

Post your code, based on what you've posted hard to say.

Re: Proc Rank

newguy1580 — Sun, 16 Mar 2014 03:04:48 GMT

proc rank data=temp4 out=temp4 groups=4;

var hisugbev;

ranks hisugbev4;

run;

proc phreg;

class hisugbev4 (ref='0');

model tpyrs*c5_1(0)= age hisugbev4/rl ties=efron;

run;

Re: Proc Rank

newguy1580 — Sun, 16 Mar 2014 03:05:25 GMT

In this case, I just get level 2 and level 3 parameter estimates and for some reason no level 1

Re: Proc Rank

PGStats — Sun, 16 Mar 2014 03:16:00 GMT

Could it be that all level=1 cases are censored? - PG

Re: Proc Rank

newguy1580 — Sun, 16 Mar 2014 03:17:11 GMT

Let me check my output and get back to you on that

Re: Proc Rank

newguy1580 — Sun, 16 Mar 2014 03:19:43 GMT

I have 36,945 total observations and 36,700 were censored leaving 245 actual cases, i personally don't think that's the reason

Re: Proc Rank

PGStats — Sun, 16 Mar 2014 03:32:45 GMT

Try running

proc freq data=temp4 ;

table hisugbev4*c5_1;

run;

Re: Proc Rank

Tom — Sun, 16 Mar 2014 03:33:08 GMT

What is the cross tab of censor and your new 4 level ranking?

Re: Proc Rank

Reeza — Sun, 16 Mar 2014 03:34:09 GMT

Did you want to include referential coding?

You may also want to point explicitly to data=temp4

proc phreg data=temp4;

class hisugbev4 (ref='0')/param=REF;

model tpyrs*c5_1(0)= age hisugbev4/rl ties=efron;

run;

what happens when you run a proc freq, none of the cells are 0?

proc freq data =temp4;

table hisugbev4*c5_1;

run;

Re: Proc Rank

newguy1580 — Sun, 16 Mar 2014 03:38:56 GMT

When I run proc freq, for level=0 (reference) I have a total of 16,875 people, level=2 11,014 and level=3 9,086 for a grand total of 36,975...so level=1 doesn't show up. I guess there was no one that had a value that fell into that particular quartile?

Re: Proc Rank

newguy1580 — Sun, 16 Mar 2014 03:42:30 GMT

I tried putting in data=temp4 next to the proc phreg statement and ran the model again but nothing different..i get the same output.

Re: Proc Rank

PGStats — Sun, 16 Mar 2014 03:44:45 GMT

You must have a large fraction of hisugbev values that are tied... You should reconsider the way you divide up that variable if you want to get a meaningful model.

Re: Proc Rank

newguy1580 — Sun, 16 Mar 2014 03:46:02 GMT

So are you saying try quintiles or tertiles instead of quartiles?

Re: Proc Rank

Reeza — Sun, 16 Mar 2014 03:49:52 GMT

Take a look at your actual values in a histogram to decide a distribution. Sometimes there's genuine data intervals visible.

Make sure to add the param=ref option as well, sorry to repeat that.

Re: Proc Rank

newguy1580 — Mon, 17 Mar 2014 00:31:32 GMT

I ran proc univariate for the variable hisugbev and it is very left-skewed, about 65% of the observations have a value of 0 so that would probably explain the missing parameter estimate. My professor said I can still create quartiles even with the skewed data. Is there any sort of ideas you may have in mind? The data goes from 0-18 so i thought i could do maybe like 0-4, 4-8, 8-12, 12+

Re: Proc Rank

PGStats — Mon, 17 Mar 2014 01:36:51 GMT

It would probably make sense. To be sure, try looking at the distribution of uncensored events:

ods graphics on;

Proc univariate data=temp4;

where c5_1 ne 0;

var hisugbev;

histogram;

run;

Re: Proc Rank

newguy1580 — Mon, 17 Mar 2014 01:49:37 GMT

The histogram is highly left-skewed, basically the same as the one after running proc univariate for hisugbev. Do you think then what I suggested would still be a reasonable idea?

Re: Proc Rank

PGStats — Mon, 17 Mar 2014 02:01:24 GMT

I would simply make sure I have a fair share of uncensored events in each category.

Good luck!