SPSSX Discussion - random number generation

3 posts

Hi,

I want to generate a set of 250 numbers drawn from the set of integers
[1,2,3,4,5,6,7}. I wonder if there is a simple way to do this in syntax, as
I haven't been able yet to do this with the various RAND. functions, other
than generating real numbers and re-computing them with other functions.

Thanks in advance for help,

Clive

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

3470 posts

* Use RV.UNIFORM function to generate integers in the range 1-7.
* From the FM:  "The uniform distribution takes values in the range a<x<b".
* Let a=1 and b=8, and then use TRUNC.

new file.
dataset close all.
input program.
loop #i = 1 to 250.
- compute X = trunc(rv.uniform(1,8)).
- end case.
end loop.
end file.
end input program.
frequencies X.


<quote author="Clive">
Hi,

I want to generate a set of 250 numbers drawn from the set of integers
[1,2,3,4,5,6,7}. I wonder if there is a simple way to do this in syntax, as
I haven't been able yet to do this with the various RAND. functions, other
than generating real numbers and re-computing them with other functions.

Thanks in advance for help,

Clive

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

3809 posts

An approach I've adopted lately.

MATRIX.
SAVE (TRUNC(UNIFORM(250,1)*7+1 ))/OUTFILE */VAR UNI7.
END MATRIX.

Bruce Weaver wrote

* Use RV.UNIFORM function to generate integers in the range 1-7.
* From the FM:  "The uniform distribution takes values in the range a<x<b".
* Let a=1 and b=8, and then use TRUNC.

new file.
dataset close all.
input program.
loop #i = 1 to 250.
- compute X = trunc(rv.uniform(1,8)).
- end case.
end loop.
end file.
end input program.
frequencies X.


<quote author="Clive">
Hi,

I want to generate a set of 250 numbers drawn from the set of integers
[1,2,3,4,5,6,7}. I wonder if there is a simple way to do this in syntax, as
I haven't been able yet to do this with the various RAND. functions, other
than generating real numbers and re-computing them with other functions.

Thanks in advance for help,

Clive

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

... [show rest of quote]

3 posts

In reply to this post by Clive

Hi Bruce

Thank you very much for your solution to this problem,

Regards,

Clive

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

3 posts

In reply to this post by Clive

Hi David

Thank you very much for your matrix solution to this problem,

Regards

Clive

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

198 posts

In reply to this post by Bruce Weaver

Hi all,

This code works great.

input program.
loop #i = 1 to 250.
- compute X =rv.univorm (.5,16).
- end case.
end loop.
end file.
end input program.
frequencies X.

Is there a way to tweak this so that it only generates numbers that 0.5 apart?

I need numbers in this range but that only jump by 0.5 so: 0.50, 1.0, 1.5, 2.0, .....15.5, 16.0.

Thanks much!
Carol

726 posts

One way to do it, truncate to integers over 1-32 [note exactly 33 should never happen in rv.uniform(1,33)], and then divide by 2. Upped the number of simulated draws to show it is approximately is uniform as expected.

*************************.
input program.
loop #i = 1 to 1e6.
- compute X =trunc(rv.univorm (1,33))/2.
- end case.
end loop.
end file.
end input program.
frequencies X.
*************************.

198 posts

Awesome andy!

This worked like a charm.

Thanks for the quick response!

3470 posts

In reply to this post by Andy W

Here is the same approach but using David's MATRIX method:

MATRIX.
SAVE (TRUNC(UNIFORM(1e6,1)*32+1)/2)/OUTFILE */VAR X.
END MATRIX.
GRAPH HISTOGRAM X.
FREQUENCIES X.

Andy W wrote

One way to do it, truncate to integers over 1-32 [note exactly 33 should never happen in rv.uniform(1,33)], and then divide by 2. Upped the number of simulated draws to show it is approximately is uniform as expected.

*************************.
input program.
loop #i = 1 to 1e6.
- compute X =trunc(rv.univorm (1,33))/2.
- end case.
end loop.
end file.
end input program.
frequencies X.
*************************.

61 posts

Well, I would amend the range "exactly 33 should never happen in rv.uniform(1,33)"
to "never in my lifetime", since the basic Mersenne generator generates in [0,1].