dear all, how to use SAS to compute empirical CDF?

[deleted_user]

I have monthly data (huge dataset), and want to compute empirical CDF for each year each company.

what's the best way to do it?


thank you so much!!!

[Cynthia_sas]

Hi:
Perhaps this documentation will be of use to you.
http://support.sas.com/rnd/app/da/new/802ce/ets/chap4/sect8.htm
in which it discusses the CDF=EMPIRICAL option.

For help using the SAS/STAT procedures, you might consider contacting Tech Support.

cynthia

[deleted_user]

thank you so much, though it is not what i am looking for. i dont think a simple build-in function can solve it. it should be a formula....i guess.....

[Cynthia_sas]

Hmmm,
I didn't think that CDF=EMPIRICAL() in the documentation example was a function. In this syntax example (clipped from the doc), it looks, to me, like a statement option on the ERRORMODEL statement.
[pre]

proc model data=t time=t itprint;
dependent y;
parm a 5 ;
y = a;
obj = resid.y * resid.y;
errormodel y ~ general( obj )
cdf=(empirical=( tails=( t(15) percent= 5)));

fit y / outns=s out=r;
id date;
solve y / data=t(where=(date='1jun95'd ))
residdata=r sdata=s random=200 seed=6789 out=monte;
run;

[/pre]

Of course, I am not a STAT person and that was just the FIRST hit I came up with when searching the doc for the string "empirical CDF".

For more help with your question, you might consider contacting Tech Support.

cynthia

[deleted_user]

sorry that i may have confused you. my question is, given a list of numbers as below, how to compute CDFs based on last 100 numbers.


0.3
1.2
2.5
0.1
0.1
0.1
0.2
0.33
0.21
.2
0.1
.3
.5
.6
.2
............................(a long long list)
etc. Message was edited by: zinita