Error when printing estimated proportions with no domains

[michaelwalshe]

Hi,

I've encountered the following error when printing a proportion estimator:

from samplics import TaylorEstimator
from samplics.datasets import load_nhanes2
from samplics.utils.types import PopParam

nhanes2 = load_nhanes2()["data"]

prop_est = TaylorEstimator(PopParam.prop)

prop_est.estimate(
    y=nhanes2["highlead"],
    psu=nhanes2["psuid"],
    stratum=nhanes2["stratid"],
    remove_nan=True,
)

print(prop_est)

Gives the following:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
File c:\\Users\\Michael.Walshe.AMADEUS\\source\\katalyze-data\\CAMIS\\testing.py:1
----> 1 print(prop_est)

File c:\\Users\\Michael.Walshe.AMADEUS\\source\\katalyze-data\\CAMIS\\renv\\python\\virtualenvs\\renv-python-3.11\\Lib\\site-packages\\samplics\\estimation\\expansion.py:109, in _SurveyEstimator.__str__(self)
    107 upper_ci = list()
    108 coef_var = list()
--> 109 for domain in self.domains:
    110     domains += np.repeat(domain, len(self.point_est[domain])).tolist()
    111     levels += list(self.point_est[domain].keys())

TypeError: iteration over a 0-d array

I think this is due to the __str__ method of SurveyEstimator checking that self.domain is not None rather than self.domain.shape in ((), (0,)) which is what it looks like the checks are elsewhere.