Various random variates may be generated by the VAR operation by means of the statistical library functions of editorial computing. (See ARIT?) To generate a random value from a distribution with the inverse distri- bution function G(), use G(rnd(0)) in the VAR operation. For example, standard normal random variates are obtained by N.G(0,1,rnd(0)) or faster by probit(rnd(0)). Other alternatives are: BIN.G(n,p,rnd(0)) binomial distribution Bin(n,p) Poisson.G(a,x) Poisson distribution Poisson(a) t.G(n,rnd(0)) t distribution with n degrees of freedom CHI2.G(n,rnd(0)) Chi-square distribution with n degrees of freedom F.G(m,n,rnd(0)) F distribution with m,n degrees of freedom For serious applications the alternative random number generator rand() is recommended instead of rnd(). R = Alternative random number generators V = More information on VAR operation M = MNSIMUL for fast generation of multivariate normal samples T = TRANSFORM <data> BY #UNIFORM(a,b) X = RNDTEST for testing sequences of random numbers

More information on Survo from www.survo.fi

Copyright © Survo Systems 2001-2012.

webmaster'at'survo.fi

Copyright © Survo Systems 2001-2012.

webmaster'at'survo.fi