predictOOS when mean_factor=False

[PhilippHoefler]

Hi,

first thank you for providing this great package!
I have been playing around with the example code and noticed that the length of the array returned by predictOOS does not match the length of the data when setting mean_factor to False.

One can use the Grunfeld data to see this:

regr = InstrumentedPCA(n_factors=1, intercept=True)
regr = regr.fit(X=data_IS, y=y_IS)
Ypred1 = regr.predictOOS(X=data_OOS, y=y_OOS, mean_factor=True)
Ypred2 = regr.predictOOS(X=data_OOS, y=y_OOS, mean_factor=False)

print(y_OOS, "\n")
print(Ypred1, "\n")
print(Ypred2, "\n")

This is what the output looks like:

firm  year
11    1954    1486.700
14    1954     459.300
10    1954     189.600
8     1954     172.490
7     1954      81.430
13    1954     135.720
15    1954      89.510
16    1954      68.600
12    1954      49.340
9     1954       5.120
6     1954       6.281
Name: invest, dtype: float64 

[780.89716506 291.46947239 380.59479685 101.20025869  65.84867622
 126.53074081  36.65394197 160.14977461  72.50646239   7.91635846
   8.04336863] 

[1244.4912375] 

Apparently the method only returns a prediction for the first entity-time pair.

[lbybee]

Hi,

Going through this now, the current implementation when mean_factor=False is only meant to produce predictions for period t+1.

There isn't really any good reason for this limitation though and it should be an easy addition to handle multiple t.

We can see about adding this soon.

[rogier117]

Hi,
I have the same issue as @PhilippHoefler. However, as mentioned in the example given by Philipp, this prediction concerns only one time period with multiple entities/firms. Looking through the code, I feel like this problem is caused by the "np.diag" part on line 911 of ipca.py.
Replacing
ypred = np.diag(X.dot(self.Gamma).dot(Factor_OOS))
with
ypred = X.dot(self.Gamma).dot(Factor_OOS)
gives me the output as desired. Do you know why on that line the diagonal of (for me) a 1-dimensional array is taken (resulting in a 1-by-1 array), or whether this is a mistake?

[zichengno1]

My solution is basically the same as @rogier117. I used ypred = np.squeeze(X.dot(self.Gamma).dot(Factor_OOS)), which is just the same as simply removing the np.diag()