Sampling posterior predictive from shifted Lognormal

[jeti]

This works (almost)

@jeti

class ShiftedLognormal(pymc3.Lognormal):
def __init__(self, mu, sigma, shift, tau=None, sd=None, *args, **kwargs):
    transform = pymc3.distributions.transforms.lowerbound(shift)
    super().__init__(mu=mu, sd=sigma, tau=tau,  *args, **kwargs,
                     transform=transform)
    self.shift = shift
    self.mean += shift
    self.mode += shift
def random(self, point, size):
    shift = draw_values([self.shift], point=point, size=size)
    return super().random(point=point, size=size) + shift

def logp(self, x):
    return super().logp(x - self.shift)

It works but the PP has shape (N_SAMPLE, 1, N_OBS) … Where is the additional dimension in the center coming from?

Thanks & regrads