Add AR model for interrupted time series

[drbenvincent]

2 added a very simple interrupted time series example with no predictors.

But it would be good to add another example where there is more temporal structure. This would then we well suited for an actual time series model, here an AR model.

• data generating function, generate_time_series_data (rename this)
• create a new AutoRegressive subclass of CausalBase

TODO

• [x] Improve interrupted time series dataset by adding temporal structure
• [x] Add another dataset with seasonality
• [ ] Implement with scikit-learn or sktime model. But pmdarima actually looks very promising. It wraps statsmodels but provides the fit/predict API.
• [ ] Implement with pymc model

[drbenvincent]

Now we have a sklearn.linear_model.LinearRegression model. When we add month as a categorical variable, this does a good job of fitting the seasonality component.

But we do still need a proper time series model here.

[drbenvincent]

Just dropping in a link to the developments on state space models by @jessegrabowski here https://discourse.pymc.io/t/state-space-models-in-pymc/9146