AttributeError: 'pandas._libs.properties.CachedProperty' object has no attribute 'func'QST:

[threehappy]

Research

• I have searched the [pandas] tag on StackOverflow for similar questions.

• I have asked my usage related question on StackOverflow.

Link to question on StackOverflow

AttributeError: 'pandas._libs.properties.CachedProperty' object has no attribute 'func'QST:

Question about pandas

%matplotlib inline

import numpy as np
import pandas as pd
import pandas_datareader.data as web

from pyfinance.ols import PandasRollingOLS

replaces pyfinance.ols.PandasRollingOLS (no longer maintained)

from statsmodels.regression.rolling import RollingOLS
import statsmodels.api as sm
from talib import RSI, BBANDS, MACD, NATR, ATR

from sklearn.feature_selection import mutual_info_classif, mutual_info_regression

import matplotlib.pyplot as plt
import seaborn as sns

---

--> 516 @appender(RegressionResults.aic.func.doc)
517 def aic(self):
518 return self._wrap(RegressionResults.aic.func(self))
520 @cache_readonly
521 @appender(RegressionResults.bic.func.doc)
522 def bic(self):

AttributeError: 'pandas._libs.properties.CachedProperty' object has no attribute 'func'

[twoertwein]

func was renamed to fget in #43828 (since 1.4) to be more consistent with python's property. I assume that the external pandas packages you are importing pandas_datareader and/or pyfinance haven't yet been tested against pandas 1.4. Unfortunately, I forgot to add a whatsnewentry for this change in #43828.

[mroeschke]

It appears that this an downstream issue in the packages mentioned since those libraries are using private APIs so they can be subject to change. I would recommend opening up an issue in the libraries you used to address the pandas usages.

[matteogabba]

After reinstalling a fresh virtual env with the latest version of statsmodels, I solved the problem.
I noticed though that the latest version of statsmodel is not available on the conda channels.

Thanks for your help!

Matteo