Prev: last negative cell
Next: TextBox lock
From: Joe User on 25 Apr 2010 18:54 Material typo.... I wrote: > I am also bothered by the likelihood of duplicate results from RANDBETWEEN > due to quantization. But I can say with impunity that that is a real > problem. I meant to write: "I can__not__ say with impunity ...". Klunk! ----- original message ----- "Joe User" <joeu2004> wrote in message news:ukhd2CM5KHA.420(a)TK2MSFTNGP02.phx.gbl... > "James Silverton" <not.jim.silverton(a)verizon.net> wrote: >> have a helper column, put randbetween(1, #values) in it > > Using RANDBETWEEN incurs a lot of unnecessary overhead and the risk of a > #NAME error and confusion if the user is not familiar with the ATP (before > Excel 2007). > > It is sufficient to fill the column with =RAND(). > > (I am also bothered by the likelihood of duplicate results from > RANDBETWEEN due to quantization. But I can say with impunity that that is > a real problem.) > > > ----- original message ----- > > "James Silverton" <not.jim.silverton(a)verizon.net> wrote in message > news:uMb6$xL5KHA.5808(a)TK2MSFTNGP02.phx.gbl... >> random wrote on Sun, 25 Apr 2010 13:37:01 -0700: >> >>> does any know if sampling from a set can be done until all >>> values are selected? >> One way would be to have a helper column, put randbetween(1, #values) in >> it, replace the random numbers with values (paste special), sort and then >> use the rest of the values sequentially. >> >> -- >> >> James Silverton >> Potomac, Maryland >> >> Email, with obvious alterations: not.jim.silverton.at.verizon.not >
From: James Silverton on 26 Apr 2010 16:54
Joe wrote on Sun, 25 Apr 2010 15:54:56 -0700: > I wrote: >> I am also bothered by the likelihood of duplicate results >> from RANDBETWEEN due to quantization. But I can say with >> impunity that that is a real problem. > I meant to write: "I can__not__ say with impunity ...". Klunk! > ----- original message ----- > "Joe User" <joeu2004> wrote in message > news:ukhd2CM5KHA.420(a)TK2MSFTNGP02.phx.gbl... >> "James Silverton" <not.jim.silverton(a)verizon.net> wrote: >>> have a helper column, put randbetween(1, #values) in it >> >> Using RANDBETWEEN incurs a lot of unnecessary overhead and >> the risk of a #NAME error and confusion if the user is not >> familiar with the ATP (before Excel 2007). >> >> It is sufficient to fill the column with =RAND(). >> >> (I am also bothered by the likelihood of duplicate results >> from RANDBETWEEN due to quantization. But I can say with >> impunity that that is a real problem.) >> >> ----- original message ----- >> >> "James Silverton" <not.jim.silverton(a)verizon.net> wrote in >> message news:uMb6$xL5KHA.5808(a)TK2MSFTNGP02.phx.gbl... >>> random wrote on Sun, 25 Apr 2010 13:37:01 -0700: >>> >>>> does any know if sampling from a set can be done until all >>>> values are selected? >>> One way would be to have a helper column, put randbetween(1, >>> #values) in it, replace the random numbers with values >>> (paste special), sort and then use the rest of the values >>> sequentially. >>> With reference to randomizing the rows of a data set, which is essentially what I suggested, it might be useful to include another helper column with integers 1,2,3.....before the random column. so that you can get the original order back again. -- James Silverton Potomac, Maryland Email, with obvious alterations: not.jim.silverton.at.verizon.not |