calerf

calerf — computes error functions.

Parameters

`x`	: real vector or matrix
`flag`	: integer indicator
`y`	: real vector or matrix (of same size than x)

calerf(x,0) computes the error function erf(x)

calerf(x,1) computes the complementary error function erfc(x)

calerf(x,2) computes the scaled complementary error function erfcx(x)



deff('y=f(t)','y=exp(-t^2)');
calerf(1,0)
2/sqrt(%pi)*intg(0,1,f)

W. J. Cody (code from Netlib (specfun))