Bug : fixef() on IV

[SwannChelly]

Hi all,

I tried to get the fixed effect estimated through the second stage of my IV and I got the following error.

---

AttributeError Traceback (most recent call last)
Cell In[24], line 1
----> 1 second_stage.fixef()

1814 _method = self._method
1815 _fml = self._fml
-> 1816 _data = self._data
1817 _weights_sqrt = np.sqrt(self._weights).flatten()
1819 blocked_transforms = ["i(", "^", "poly("]

AttributeError: 'Feiv' object has no attribute '_data'

I think that the function fixef does not support Feiv. Apologies if I made a mistake in the usage of such function.

Thank you for your involvement.

Best,
Swann

[s3alfisc]

Hi Swann, thanks for reporting! Indeed that's not supported - I just didn't consider anyone could need this / didn't see a use case. Could you provide me with an example? What are you using this for? I think it would be possible to implement support, but not super straightforward.

[SwannChelly]

Hi,

Thank you for the quick answer. I was doing some pass-through analysis with an instrumental variable strategy and I was wondering if the pass through of the common shock was similar than the pass through of idiosyncratic shocks. Since I knew fixef was working for feols, it was a quickest way for me to get the estimation of the fixed effect (I know I was lazy).

In the end I coded it myself. I totally understand that it is a minor issue and I don't really need it to be fixed, but I was just pointing out for you to have in mind the issue.

Best,
Swann