Dear all, Does anyone now how the scale factor for marginal effects is calculated in a random parameters ordered probit model? Is it N01(b'x) ? I'm trying to replicate it with no success so far. Kind regards, Andreas