def test_model_evaluation(self):
models = [ARIMA((1, 1, 0)), ARIMA((2, 1, 1)), VAR(1), VAR(3)]
results = model_evaluation(data, models, n_folds=10, forecast_steps=3)
results = np.array(results, dtype=object)
self.assertEqual(results.shape, (4 * 2 + 1, 8))
np.testing.assert_equal(results[1:, 1:].astype(float) > 0, True)
if var.is_continuous and var != data.time_variable
])
@Inputs.forecast
def set_forecast(self, forecast, id):
if forecast is not None:
self.forecasts[id] = forecast
else:
self.forecasts.pop(id, None)
# TODO: update currently shown plots
if __name__ == "__main__":
from AnyQt.QtWidgets import QApplication
from orangecontrib.timeseries import ARIMA, VAR
a = QApplication([])
ow = OWLineChart()
airpassengers = Timeseries('airpassengers')
ow.set_data(airpassengers),
msft = airpassengers.interp()
model = ARIMA((3, 1, 1)).fit(airpassengers)
ow.set_forecast(model.predict(10, as_table=True), 0)
model = VAR(4).fit(msft)
ow.set_forecast(model.predict(10, as_table=True), 1)
ow.show()
a.exec()
def test_predict_as_table(self):
model = VAR(2)
model.fit(data)
forecast = model.predict(10, as_table=True)
self.assertEqual(len(forecast.domain.variables), 2 * (1 + 2))
def test_predict(self):
model = VAR(2)
model.fit(data)
forecast, ci95_low, ci95_high = model.predict(10)
self.assertTrue(
np.logical_and(forecast > ci95_low, forecast < ci95_high).all())
if __name__ == "__main__":
from AnyQt.QtWidgets import QApplication
from Orange.data import Domain
from orangecontrib.timeseries import ARIMA, VAR
a = QApplication([])
ow = OWModelEvaluation()
data = Timeseries('yahoo_MSFT')
# Make Adjusted Close a class variable
attrs = [var.name for var in data.domain.attributes]
if 'Adj Close' in attrs:
attrs.remove('Adj Close')
data = Timeseries(
Domain(attrs, [data.domain['Adj Close']], None,
source=data.domain), data)
ow.set_data(data)
ow.set_model(ARIMA((1, 1, 1)), 1)
ow.set_model(ARIMA((2, 1, 0)), 2)
# ow.set_model(ARIMA((0, 1, 1)), 3)
# ow.set_model(ARIMA((4, 1, 0)), 4)
ow.set_model(VAR(1), 11)
ow.set_model(VAR(5), 12)
# ow.set_model(VAR(6), 14)
ow.show()
a.exec()
super().closeEvent(event)
def closeContext(self) -> None:
"""
Gather configs in contextVariables and close context.
"""
if not self.features:
# only close in case of when features are not present if they are
# feature selection is defined by the input and context should
# not have impact
attrs, is_logit = [], []
for config in self.configs:
attrs.append(config.get_selection())
is_logit.append(config.is_logarithmic)
self.attrs = attrs
self.is_logit = is_logit
super().closeContext()
if __name__ == "__main__":
from orangecontrib.timeseries import ARIMA, VAR
airpassengers = Timeseries.from_file('airpassengers')
msft = airpassengers.interp()
model1 = ARIMA((3, 1, 1)).fit(airpassengers)
model2 = VAR(4).fit(msft)
ow = WidgetPreview(OWLineChart)
ow.run(set_data=airpassengers,
set_forecast=[(model1.predict(10, as_table=True), 0),
(model2.predict(10, as_table=True), 1)])
def create_learner(self):
ic = self.IC_LABELS[tuple(self.IC_LABELS.keys())[self.ic]]
trend = self.TREND_LABELS[tuple(self.TREND_LABELS.keys())[self.trend]]
return VAR(self.maxlags, ic, trend)