topic Re: Probability that the sample means are between 2 and 3? uniform distribution. in SAS Programming

Probability that the sample means are between 2 and 3? uniform distribution.

valarievil — Sat, 19 Oct 2019 18:43:19 GMT

I have a large set of random uniform distributed numbers between -6 and 10. My goal is to find the probability that the mean of my sample is between 2 and 3. I used the advice from a related post (https://communities.sas.com/t5/General-SAS-Programming/Random-numbers-Sample-size-then-calculate-means-and-Standard/m-p/403537/highlight/true#M50751)

But I am still getting an error, Here is the code I am attempting:

data Prob1A;
call streaminit (0);
k= 2000;
n= 40000;
do i=1 to k;
	xuni= rand('uniform', -6, 10);
	output;
keep xuni; 
end;
proc means data=Prob1A noprint;
by SampleID;
var xuni;
output out=1Ameans mean=SampleMean std=SampleStd;
run;
data p23;
set 1Ameans;
A1sol = (2< SampleMean < 3);
run;
proc freq data= p23;
tables A1sol;
run;

I'm not getting any output except for the error message: ERROR: Variable A1SOL not found. Why does my program not recognize this variable A1SOL? I thought it was adequately defined. Also I'm a bit stuck on how to proceed from here. I would like to be able to print the probability that the sample means are between 2 and 3. Thanks in advance!

Re: Probability that the sample means are between 2 and 3? uniform distribution.

VDD — Sat, 19 Oct 2019 19:19:19 GMT

because you proc means fails.

this by SampleID;

you have to have the by variable in the dataset being evaluated.

Re: Probability that the sample means are between 2 and 3? uniform distribution.

PaigeMiller — Sat, 19 Oct 2019 19:26:35 GMT

@valarievil wrote:

I have a large set of random uniform distributed numbers between -6 and 10. My goal is to find the probability that the mean of my sample is between 2 and 3. I used the advice from a related post (https://communities.sas.com/t5/General-SAS-Programming/Random-numbers-Sample-size-then-calculate-means-and-Standard/m-p/403537/highlight/true#M50751)

But I am still getting an error, Here is the code I am attempting:
data Prob1A;
call streaminit (0);
k= 2000;
n= 40000;
do i=1 to k;
	xuni= rand('uniform', -6, 10);
	output;
keep xuni; 
end;
proc means data=Prob1A noprint;
by SampleID;
var xuni;
output out=1Ameans mean=SampleMean std=SampleStd;
run;
data p23;
set 1Ameans;
A1sol = (2< SampleMean < 3);
run;
proc freq data= p23;
tables A1sol;
run;
I'm not getting any output except for the error message: ERROR: Variable A1SOL not found. Why does my program not recognize this variable A1SOL? I thought it was adequately defined. Also I'm a bit stuck on how to proceed from here. I would like to be able to print the probability that the sample means are between 2 and 3. Thanks in advance!

Surely there must be an earlier error in the LOG. Your PROC MEANS must have an error as well.

Re: Probability that the sample means are between 2 and 3? uniform distribution.

valarievil — Sat, 19 Oct 2019 19:33:13 GMT

What should I use instead of sampleID?

Re: Probability that the sample means are between 2 and 3? uniform distribution.

valarievil — Sat, 19 Oct 2019 19:35:29 GMT

But why is the variable A1sol not defined?

Re: Probability that the sample means are between 2 and 3? uniform distribution.

VDD — Sat, 19 Oct 2019 19:35:57 GMT

@valarievil you can use SampleID you just have to have it in the SAS dataset. So your first error in the process was at your proc means because the incoming dataset did not have SampleID in it.

Re: Probability that the sample means are between 2 and 3? uniform distribution.

VDD — Sat, 19 Oct 2019 20:03:51 GMT

Please slow down.

you need to get a SampleID variable in this process before you can continue.

proc means data=Prob1A noprint;
by SampleID;
var xuni;
output out=1Ameans mean=SampleMean std=SampleStd;
run;

data Prob1A;
call streaminit (0);
k= 2000;
n= 40000;
do i=1 to k;
xuni= rand('uniform', -6, 10);
output;
keep xuni;
end;

Where is the SampleID that proc means is doing the by on in Prob1A????

data Prob1A;
retain SampleID 0;
call streaminit (0);
k= 200;
n= 4000;
do b = 1 to 10;
do i=1 to k;
SampleID =b +1;
xuni= rand('uniform', -6, 10);
output;
keep xuni SampleID;
end;
end;
run;

Re: Probability that the sample means are between 2 and 3? uniform distribution.

valarievil — Sat, 19 Oct 2019 19:55:31 GMT

I keep getting the error "variable A1SOL not found". Why?

Re: Probability that the sample means are between 2 and 3? uniform distribution.

VDD — Sat, 19 Oct 2019 20:02:35 GMT

you have a sas dataset that has a 1 at the begging.

remove the 1 from 1Ameans because that dataset is really Ameans when created by SAS

data p23;
set 1Ameans; A1sol = (2< SampleMean < 3);
run;
proc freq data= p23;
tables A1sol;
run;

Re: Probability that the sample means are between 2 and 3? uniform distribution.

valarievil — Sat, 19 Oct 2019 20:03:32 GMT

Will somebody please tell me why I keep getting the error "variable A1SOL not found"? What did I do wrong?

Re: Probability that the sample means are between 2 and 3? uniform distribution.

PaigeMiller — Sat, 19 Oct 2019 20:04:36 GMT

People have told you what you did wrong. You have errors earlier in the program, as PROC MEANS cannot work. I believe @VDD has pointed them out.

Re: Probability that the sample means are between 2 and 3? uniform distribution.

VDD — Sat, 19 Oct 2019 20:21:18 GMT

then you have an incorrect SAS dataset name

calling it 1A can not happen SAS droped the 1 when it created the dataset here

proc means data=Prob1A noprint;
by SampleID;
var xuni;
output out=1Ameans mean=SampleMean std=SampleStd;
run;

And you need to do a proc sort on the incoming dataset if you use the by statement.

Did you @valarievil need to call the class SampleID or the by SampleID statement here.

Re: Probability that the sample means are between 2 and 3? uniform distribution.

valarievil — Sat, 19 Oct 2019 20:57:24 GMT

I tried that and I'm still not getting an output.

Re: Probability that the sample means are between 2 and 3? uniform distribution.

VDD — Sat, 19 Oct 2019 22:38:05 GMT

please show complete log for ever datastep and proc somethings are issue prior to the final error. Start at the begging.

Re: Probability that the sample means are between 2 and 3? uniform distribution.

valarievil — Sun, 20 Oct 2019 18:08:20 GMT

Thanks for everyone's responses! After accumulating your collective advice, I have come up with a code that gives me what I need:

data Prob1A;
call streaminit (47);
k= 2000;
n= 40000;
do sampleid=1 to k;
 do j=1 to n;
	xuni= rand('uniform', -6, 10);
	output;
 end;
end;
keep sampleid xuni; 
run;
proc means data=Prob1A noprint;
by SampleID;
var xuni;
output out=A1means mean=SampleMean std=SampleStd;
run;
data p197_24;
set A1means;
A1sol = (1.97< SampleMean < 2.4);
run;
proc freq data= p197_24;
tables A1sol;
run;

Thanks again!

Sorry the parameters are different than the original question.