PandasRollingOLS reports ValueError

[zhangda425]

the following code cannot work and report ValueError. hope to see a better version soon, thx.

from pyfinance import ols
data = {'A':[2,3,4,5,6],'B':[10,11,12,13,14]}
df = pd.DataFrame(data)
rolling = ols.RollingOLS(y=df['B'], x=df['A'], window=3,has_const=False,use_const=False)

File "C:\ProgramData\Anaconda3\lib\site-packages\pyfinance\utils.py", line 705, in rolling_windows
' a, {1}.'.format(window, a.shape[0]))

ValueError: Specified window length of 3 exceeds length of a, 1.

[zhangda425]

I fix the bug above by adding
if x.shape[0]==1:
x=x.T
into the function _clean_xy. But when I call the property rsq of rolling, there bumps another bug.

from pyfinance import ols
data = {'A':range(0,500),'B':range(500,1000)}
df = pd.DataFrame(data)
y=df['B']
x=df['A']
rolling = ols.PandasRollingOLS(y=y, x=x, window=3,has_const=False,use_const=False)
rolling.rsq
Traceback (most recent call last):

File "", line 1, in
rolling.rsq

File "C:\ProgramData\Anaconda3\lib\site-packages\pyfinance\ols.py", line 822, in rsq
return self._wrap_series(stat='_rsq')

File "C:\ProgramData\Anaconda3\lib\site-packages\pyfinance\ols.py", line 749, in _wrap_series
return Series(getattr(self, stat), index=self.ridx, name=name)

File "C:\ProgramData\Anaconda3\lib\site-packages\pyfinance\ols.py", line 471, in _rsq
return self._ss_reg / self._ss_tot

File "C:\ProgramData\Anaconda3\lib\site-packages\pyfinance\ols.py", line 459, in _ss_reg
return np.sum(np.square(self._predicted

File "C:\ProgramData\Anaconda3\lib\site-packages\pyfinance\ols.py", line 426, in _predicted
axis=-1)))

ValueError: shapes (498,3) and (498,1) not aligned: 3 (dim 1) != 498 (dim 0)

[zhangda425]

it seems that you use matrix multiplication.  should it be dot multiplication? I don't know. it takes so much efforts to fix them........

def _predicted(self):
"""The predicted values of y ('yhat')."""
return np.squeeze(np.matmul(self.xwins, np.expand_dims(self.solution,
axis=-1)))