BUG: Get prediction index in VAR predict.

[ChadFulton]

Closes #4784

See also 6c1f892#diff-e2d6024b9d0ed6c0410bf2fffc1f1254R342

[ChadFulton]

Another issue that came up: predict also breaks if the params is not a numpy array, e.g. if it is a Pandas dataframe (which is what the res.params is if the passed endog was a Pandas dataframe). e.g.:

import pandas as pd
import statsmodels.api as sm

dta = sm.datasets.macrodata.load_pandas().data
dta.index = pd.PeriodIndex(start='1959Q1', end='2009Q3', freq='Q')

endog = dta[['realgdp', 'realcons', 'realinv']]

mod = sm.tsa.VAR(endog)
res = mod.fit(maxlags=1)

mod.predict(res.params, start=0, end=500)

yields:

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-1-a2b015a59706> in <module>()
     10 res = mod.fit(maxlags=1)
     11 
---> 12 mod.predict(res.params, start=0, end=500)

~/projects/statsmodels/statsmodels/tsa/vector_ar/var_model.py in predict(self, params, start, end, lags, trend)
    563         # fit out of sample
    564         y = y[-k_ar:]
--> 565         coefs = params[k_trend:].reshape((k_ar, k, k)).swapaxes(1, 2)
    566         predictedvalues[pv_end:] = forecast(y, coefs, intercept, out_of_sample)
    567         return predictedvalues

~/miniconda3/envs/python3/lib/python3.6/site-packages/pandas/core/generic.py in __getattr__(self, name)
   4370             if self._info_axis._can_hold_identifiers_and_holds_name(name):
   4371                 return self[name]
-> 4372             return object.__getattribute__(self, name)
   4373 
   4374     def __setattr__(self, name, value):

AttributeError: 'DataFrame' object has no attribute 'reshape'

This issue is fixed in 322e743

[coveralls]

Coverage decreased (-0.03%) to 83.589% when pulling 322e743 on ChadFulton:var-predict into 56fa48e on statsmodels:master.

[codecov-io]

Codecov Report

Merging #4785 into master will increase coverage by <.01%.
The diff coverage is 0%.

@@            Coverage Diff             @@
##           master    #4785      +/-   ##
==========================================
+ Coverage    81.1%   81.11%   +<.01%     
==========================================
  Files         581      581              
  Lines       89787    89785       -2     
  Branches    10027    10027              
==========================================
- Hits        72826    72825       -1     
+ Misses      14682    14681       -1     
  Partials     2279     2279

Impacted Files	Coverage Δ
statsmodels/tsa/vector_ar/var_model.py	88.54% <0%> (+0.09%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 56fa48e...322e743. Read the comment docs.

[ChadFulton]

There is no coverage for predict, but at least "it's green" means I didn't break var_model.py with a syntax error. Guess we can merge whenever.

[ChadFulton]

Guess I will merge this. It's pretty straightforward.