evaluaton of RAND() function

[Stephen Cox]

Hi Folks,

does anyone know of any evaluations of the RAND function of
excel...specifically how well it works at returning an even distribution of
numbers. I know it is supposed to, but random number generators do not
always do so just because they are supposed to.

Cheers,

Stephen

 

[Jerry W. Lewis]

See McCullough and Wilson's papers in Computational Statistics and Data
Analysis 31:27-37,1999 ("On the accuracy of statistical procedures in
Microsoft Excel 97") and 40:713-721,2002 ("On the accuracy of
statistical procedures in Microsoft Excel 2000 and Excel XP"). RAND()
in Excel versions prior to 2003 is not suitable for serious simulation
work. RAND in Excel 2003 is quite good, and the algorithm is documented
http://support.microsoft.com/default.aspx?scid=kb;en-us;828795&Product=xl2003
or you could implement the Mersenne Twister

http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/emt.html
http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/VERSIONS/FORTRAN/fortran.html
http://www-personal.engin.umich.edu/~wagnerr/MersenneTwister.html

If you comply with their license, you could use the freeware
implementation in NtRand

http://www.numtech.com/NtRand/

Jerry

 

[Stephen Cox]

hey thanks very much - this is exactly what i was looking for!!

Cheers,

Stephen

 

[Jerry W. Lewis]

You are welcome.
Jerry