Previous: rmpl
Up: Manual Pages
Next: setlog
Previous Page: rmpl
Next Page: setlog

rndgen

• name:
  rndgen
• synopsis:

  rndgen outputfile [-number_of_points= integer] [-seed=[ integer]] [-pairs] [-test] [-uniform[= lower,upper[,n_to_sum]] | [-gaussian[= mean,sigma] [-limit_in_sigmas= value]]

• description:
  rndgen generates a data set with random numbers for the ordinates or for both columns, with gaussian and uniform distributions supported. The random number generator uses a random shuffling routine with two linear congruential generators.
• files:
  The new data set is placed in outputfile.
• options:

  number_of_points

  
  Specifies the number of points to generate. The default is 100.

  seed

  
  Specifies the integer seed value for the random number generator. A large odd number is adviseable. If not given, or if non-positive, a seed is generated using the system clock.

  pairs

  
  Requests that random abscissae and ordinates will be generated. By default, the abscissae are successive integers.

  test

  
  Requests that statistical tests be applied to the random numbers and reported to the terminal. The Kolmogorov-Smirnov test is applied, and the correlation between successive random numbers is computed.

  uniform

  
  Specifies the generation of uniformly-distributed random numbers on the interval [ lower, upper]. If n_to_sum is specified, then each of the output values are obtained by summing n_to_sum random numbers from a uniform distribution. The default interval is [0, 1] with no summing, but since the uniform distribution is also the default, one need never give uniform unless one needs to change the interval or the number summed.

  gaussian, limit_in_sigmas

  
  gaussian requests generation of random numbers from a gaussian distribution with the specified mean and sigma; the defaults are 0 and 1, respectively. limit_in_sigmas specifies the maximum acceptable deviation of a random number from the mean in units of sigma; by default there is no limit.