Length of endogenous variable must be larger the the number of lags used

Question

I was recently following this tutorial on Time Series Analysis in Python by Susan Li. I am fitting a time series SARIMAX model on the following series:

y['2017':]

OUT: 
Order Date
2017-01-01     397.602133
2017-02-01     528.179800
2017-03-01     544.672240
2017-04-01     453.297905
2017-05-01     678.302328
2017-06-01     826.460291
2017-07-01     562.524857
2017-08-01     857.881889
2017-09-01    1209.508583
2017-10-01     875.362728
2017-11-01    1277.817759
2017-12-01    1256.298672
Freq: MS, Name: Sales, dtype: float64

using the following:

mod = sm.tsa.statespace.SARIMAX(y,
                                order=(1, 1, 1),
                                seasonal_order=(1, 1, 0, 12),
                                enforce_stationarity=False,
                                enforce_invertibility=False)

results = mod.fit()

print(results.summary().tables[1])

Now, this works well until here, but then when I try to visualize the results, I obtain the following error:

results.plot_diagnostics(figsize=(16, 8))

OUT: 
ValueError                                Traceback (most recent call last)
<ipython-input-16-6cfeaa52b7c1> in <module>
----> 1 results.plot_diagnostics(figsize=(16, 8))
      2 plt.show()

~/opt/anaconda3/lib/python3.8/site-packages/statsmodels/tsa/statespace/mlemodel.py in plot_diagnostics(self, variable, lags, fig, figsize, truncate_endog_names)
   4282 
   4283         if resid.shape[0] < max(d, lags):
-> 4284             raise ValueError(
   4285                 "Length of endogenous variable must be larger the the number "
   4286                 "of lags used in the model and the number of observations "

ValueError: Length of endogenous variable must be larger the the number of lags used in the model and the number of observations burned in the log-likelihood calculation.

<Figure size 1152x576 with 0 Axes>

Does anyone have any idea how to fix this, if it's some kind of library error, and if it cannot directly fixed, then how can I obtain all the diagnostic plots?

Answer 1

While defining the model, remove the parameter (enforce_stationarity = False ) and it should work fine !

Answer 2

For New people what Shantanu mean is - Instead of this -

mod = sm.tsa.statespace.SARIMAX(y,
                            enter code hereorder=(1, 1, 1),
                            seasonal_order=(1, 1, 0, 12),
                            enforce_stationarity=False,
                            enforce_invertibility=False)

You can write - enforce_stationarity=False, - this is removed, you can comment it as well - #enforce_stationarity=False,

mod = sm.tsa.statespace.SARIMAX(y, order=(1, 1, 1),
                            seasonal_order=(1, 1, 0, 12),
                            enforce_invertibility=False)