terminate(modelHelipad, datapandas.DataFrame)

Notes and Examples

1. 

   charwick

   Mar 26, 2020 at 18:12

   The terminate hook is a good place to do any post-model analysis, as the data will be finalized by that point. This example runs an OLS regression on two columns of model data.

   @heli.hook
   def terminate(model, data):
   	import statsmodels.api as sm
   	reg = sm.OLS(data['M0'], data['price-level']).fit()
   	predictions = reg.predict(data['M0'])
   
   	# Print out the statistics
   	reg.summary()
2. 

   charwick

   Jul 12, 2022 at 18:53

   If you launch another Matplotlib window in post-analysis, you must make sure not to block the event loop; otherwise you cannot launch a new model until the window is closed. This can be done with the block=False argument on pyplot.show().

   This example shows an impulse response function after each model run, without blocking further runs.

   @heli.hook
   def terminate(model, data):
   	from statsmodels.tsa.api import VAR
   	import matplotlib.pyplot as plt
   
   	model = VAR(data)
   	results = model.fit(50)
   	irf = results.irf(50)
   	irf.plot(orth=False, impulse='M0', response='ratio-jam-axe')
   	plt.show(block=False)
3. 

   Contribute a Note

   Your email address will not be published. Required fields are marked *

   Comment *

   You may use limited HTML for formatting. Please embed blocks of code in <pre><code> </code></pre> tags.

   Name *

   Email *

   Website

   Δ