Which distribution to choose for randomly select calibration data indices?

Dear all:

I have one data, nx1, and I would like to randomly select some of them
as calibration data set, and the left as test data set. I think I can
use rand() or randn(). The question here I have is which one to
choose? Should I choose from uniform distribution (i.e., rand()) or
normal distribution (i.e., randn()). I guess I will need to use
uniform distribution, but would like to make sure.

Thanks for any help in advance.
.

Relevant Pages

  • RE: Find first value != to something in Excel 2003
    ... I have a data set where there might be thousands of lines with a column ... Dim rngColumnHeader as Range,rngSearchRange as Range ... Dim lngPowerSetting as Long, rngDifferent as Range ... The Power Setting column in my test data set has about 3000 ...
    (microsoft.public.excel.programming)
  • Re: Clarification in rand(seed,111)
    ... I am using rand() function for my research work in optimization ... here in this regard i want to know whether any significant meaning is ... tried using different random seeds on different sets of data? ... If for a particular data set you find that random seed 111 gives good ...
    (comp.soft-sys.matlab)
  • Find first value != to something in Excel 2003
    ... I have a data set where there might be thousands of lines with a column ... Dim rngColumnHeader as Range,rngSearchRange as Range ... Dim lngPowerSetting as Long, rngDifferent as Range ... The Power Setting column in my test data set has about 3000 ...
    (microsoft.public.excel.programming)
  • Re: Training Decision Tree
    ... Then i created a SSIS package with OLE DB Source the validation data set ... the first time (when i used the test data set), ...
    (microsoft.public.sqlserver.datamining)
  • Re: Spring 2010 MATLAB Contest, April 28 - May 5
    ... Sample data set is tranparent to all contesters for them to develop algorithms and to tuning parameters. ... Test data set is grey to contesters. ...
    (comp.soft-sys.matlab)