Thanks Nick, that worked nicely. Do you know why didn't -egen
cut(),group()- produce similar results? I thought when you specifed
group it would bin the -cut()- values into -group()- sets of equal size.
I thought perhaps there might be a problem with ties, but it seems
unlikely since the random values are floats.
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Nick Cox
Sent: Wednesday, January 05, 2005 5:05 PM
To: [email protected]
Subject: st: RE: Random numbers and egen cut()
Well, random numbers are expected to
display a certain amount of randomness...
I'd
. set obs 2400
and use -egen, seq()- before
and after
. gen r = uniform()
. sort block r
Nick
[email protected]
Wallace, John
> Happy New Year Statalist!
>
> I have 6 blocks of 400 samples that I'd like to randomly
> assign to groups of 50. I thought I could do this with egen
> cut, but I got unanticipated results
>
> .set obs 6
> .gen block = _n
> .expand(400)
> .sort v1
> .egen id = fill(1/400 1/400)
> .gen random = uniform()
> .sort block random
> .egen group = cut(random), group(8)
>
> .tab group block
>
> | block
> group | 1 2 3 4 5
> 6 | Total
> ------+-------------------------------------------------------
> -+-------
> 0 | 47 51 54 46 48
> 54 | 300
> 1 | 51 61 49 46 46
> 47 | 300
> 2 | 61 41 42 48 56
> 52 | 300
> 3 | 53 33 46 60 54
> 54 | 300
> 4 | 48 57 52 52 37
> 54 | 300
> 5 | 46 57 53 52 51
> 41 | 300
> 6 | 43 52 51 49 55
> 50 | 300
> 7 | 51 48 53 47 53
> 48 | 300
> ------+-------------------------------------------------------
> -+-------
> Total | 400 400 400 400 400
> 400 | 2,400
>
> What would have been a better way to do this?
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
