| Date: | Sat, 25 Apr 1998 12:09:16 -0600 |
|---|
| Reply-To: | jshi@KUHUB.CC.UKANS.EDU |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU> |
|---|
| From: | Jingren Shi <jshi@KUHUB.CC.UKANS.EDU> |
|---|
| Organization: | Deja News - The Leader in Internet Discussion |
|---|
| Subject: | random generation |
|---|
How to generate chi square, t and F distribution,
If you have x that is normally dritributed, you may generate chi square
distribution. If you have normall dritribution and chi square distribution,
you may generate student distribution and F distibution.
How to randomly draw a simple size 2 from 10 without replacement, The
general proof may be difficult. You may think a simple case selecting 2 out
of 3. The simulation result seems good.
data t1;
do index = 1 to 10;
var=ranuni(0);
output;
end;
run;
proc sort data = t1 out=t2;
by var;
run;
data t3;
set t2(obs=2);
run;
/* here is the simulation */
libname sd '~model03/sas';
%macro doloop;
%do i = 0 %to 10000;
data t1;
do index = 1 to 10;
var=ranuni(&i*2);
output;
end;
run;
proc sort data = t1 out=t2;
by var;
run;
data t3;
set t2(obs=2);
run;
proc append base=base data=t3; run;
%end;
%mend;
%doloop;
proc freq data=base;
table index /out=sd.out;
run;
/* here is the result */
The SAS System 00:51 Saturday, April 25, 1998 1
Cumulative Cumulative INDEX Frequency Percent Frequency Percent
--------------------------------------------------- 1 2011 10.1 2011
10.1 2 1970 9.8 3981 19.9 3 2043 10.2 6024 30.1 4 2023 10.1
8047 40.2 5 1971 9.9 10018 50.1 6 1995 10.0 12013 60.1 7 2026
10.1 14039 70.2 8 2028 10.1 16067 80.3 9 1972 9.9 18039 90.2 10
1963 9.8 20002 100.0
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading
| 
