def test_prediction(self):
data_frame = datasets.get('gas_2016_hour')
data_frame_month = data_frame.resample('MS').sum().loc['2016', :]
data_frame_training = data_frame_month.iloc[:-1, :]
data_frame_pred = data_frame_month.iloc[[-1], :]
mvlr = regression.MultiVarLinReg(data_frame=data_frame_month,
dependent_var='313b',
options={'p_max': 0.04})
mvlr.do_analysis()
data_frame_pred_95 = mvlr._predict(
mvlr.fit, data_frame=data_frame_pred)
mvlr.confint = 0.98
data_frame_pred_98 = mvlr._predict(
mvlr.fit, data_frame=data_frame_pred)
self.assertAlmostEqual(
data_frame_pred_95.loc['2016-12-01', 'predicted'], data_frame_pred_98.loc['2016-12-01', 'predicted'])
self.assertTrue(data_frame_pred_98.loc['2016-12-01', 'interval_u']
> data_frame_pred_95.loc['2016-12-01', 'interval_u'])
self.assertTrue(data_frame_pred_98.loc['2016-12-01', 'interval_l']
< data_frame_pred_95.loc['2016-12-01', 'interval_l'])
# check limitation to zero
mvlr.allow_negative_predictions = False
mvlr.add_prediction()
self.assertTrue(mvlr.data_frame['predicted'].min() >= 0)
def test_init(self):
data_frame = datasets.get('gas_2016_hour')
data_frame_month = data_frame.resample('MS').sum()
mvlr = regression.MultiVarLinReg(data_frame=data_frame_month,
dependent_var='313b',
options={'p_max': 0.04})
mvlr.do_analysis()
self.assertTrue(hasattr(mvlr, 'list_of_fits'))
def test_strange_names(self):
data_frame = datasets.get('gas_2016_hour')
data_frame_month = data_frame.resample('MS').sum()
data_frame_month.rename(columns={'d5a7': '3*tempête !'}, inplace=True)
mvlr = regression.MultiVarLinReg(data_frame=data_frame_month,
dependent_var='313b',
options={'p_max': 0.04})
mvlr.do_analysis()
self.assertTrue(hasattr(mvlr, 'list_of_fits'))
def test_plot(self):
data_frame = datasets.get('gas_2016_hour')
data_frame_month = data_frame.resample('MS').sum()
mvlr = regression.MultiVarLinReg(data_frame=data_frame_month,
dependent_var='313b',
options={'p_max': 0.04})
mvlr.do_analysis()
with mock.patch.object(plt_mocked, 'subplots', return_value=(fig_mock, ax_mock)):
mvlr.plot()
def test_predict(self):
data_frame = datasets.get('gas_2016_hour')
data_frame_month = data_frame.resample('MS').sum()
data_frame_month.rename(columns={'d5a7': '3*tempête !'}, inplace=True)
mvlr = regression.MultiVarLinReg(data_frame=data_frame_month,
dependent_var='313b',
options={'p_max': 0.04})
mvlr.do_analysis()
mvlr.add_prediction()
self.assertListEqual(mvlr.data_frame.columns.tolist(),
data_frame_month.columns.tolist() + ['predicted', 'interval_l', 'interval_u'])
def test_raises(self):
data_frame = datasets.get('gas_2016_hour')
data_frame_month = data_frame.resample('MS').sum()
mvlr = regression.MultiVarLinReg(data_frame=data_frame_month,
dependent_var='313b',
options={'p_max': 0.04})
self.assertRaises(UnboundLocalError, mvlr.add_prediction)
try:
x = mvlr.list_of_fits
self.assertTrue(False)
except UnboundLocalError:
self.assertTrue(True)
def test_prune(self):
"Create overfitted model and prune it"
data_frame = datasets.get('gas_2016_hour')
data_frame_month = data_frame.resample('MS').sum()
mvlr = regression.MultiVarLinReg(data_frame=data_frame_month,
dependent_var='313b')
mvlr.do_analysis()
self.assertTrue("ba14" in mvlr.fit.model.exog_names)
pruned = mvlr._prune(mvlr.fit, 0.05)
self.assertTrue("ba14" in pruned.model.exog_names)
# with this value, both x will be removed, which is a bit counter-intuitive because initially only ba14 has a pvalue > p_max.
pruned = mvlr._prune(mvlr.fit, 0.00009)
self.assertFalse("ba14" in pruned.model.exog_names)
self.assertFalse("d5a7" in pruned.model.exog_names)
mvlr = regression.MultiVarLinReg(data_frame=data_frame_month,
dependent_var='313b',
options={'p_max': 0.00009})
mvlr.do_analysis()
self.assertFalse("ba14" in mvlr.fit.model.exog_names)
self.assertFalse("d5a7" in mvlr.fit.model.exog_names)
def test_alternative_metrics(self):
data_frame = datasets.get('gas_2016_hour')
data_frame_month = data_frame.resample('MS').sum()
mvlr = regression.MultiVarLinReg(data_frame=data_frame_month,
dependent_var='313b',
options={'p_max': 0.04})
mvlr.do_analysis()
best_rsquared = mvlr.find_best_rsquared(mvlr.list_of_fits)
best_akaike = mvlr.find_best_akaike(mvlr.list_of_fits)
best_bic = mvlr.find_best_bic(mvlr.list_of_fits)
self.assertEqual(best_rsquared, best_akaike)
self.assertEqual(best_rsquared, best_bic)
def test_pickle_round_trip(self):
"Pickle, unpickle and check results"
data_frame = datasets.get('gas_2016_hour')
data_frame_month = data_frame.resample('MS').sum().loc['2016', :]
data_frame_training = data_frame_month.iloc[:-1, :]
data_frame_pred = data_frame_month.iloc[[-1], :]
mvlr = regression.MultiVarLinReg(data_frame=data_frame_month,
dependent_var='313b',
options={'p_max': 0.04})
mvlr.do_analysis()
data_frame_pred_95_orig = mvlr._predict(
mvlr.fit, data_frame=data_frame_pred)
s = pickle.dumps(mvlr)
m = pickle.loads(s)
self.assertTrue(hasattr(m, 'list_of_fits'))
data_frame_pred_95_roundtrip = m._predict(
m.fit, data_frame=data_frame_pred)
self.assertAlmostEqual(
data_frame_pred_95_orig.loc['2016-12-01', 'predicted'], data_frame_pred_95_roundtrip.loc['2016-12-01', 'predicted'])