Re: [Help-gsl] dawsons integral for imaginary arguments

[Brian Gough]

Carsten Fortmann writes:
 > I need the following expression
 > 
 > exp(x*x)erfc(x)  (1).
 > 
 > Due to the exp(x*x) (not exp(-x*x)!!!),
 >   this expression returns a  nan for large arguments. I found out, that 
 > (1) is in principal given by dawsons integral for purely imaginary 
 > argument. Does anyone know how to obtain this in gsl?

Hello,

Apart from constant factors it is 1/gsl_sf_hazard().
See http://www.network-theory.co.uk/docs/gslref/gsl-ref_105.html

-- 
Brian Gough

Network Theory Ltd,
Publishing the GSL Manual - http://www.network-theory.co.uk/gsl/manual/