def test_poisson_predict_length():
"""
Tests that the length of the predict dataframe is equal to no of steps h
"""
model = pf.GASX(formula="y ~ x1", data=data, ar=1, sc=1, family=pf.Poisson())
x = model.fit()
x.summary()
assert(model.predict(h=5, oos_data=data_oos).shape[0] == 5)
def test_poisson_predict_is_nans():
"""
Tests that the predictions in-sample are not NaNs
"""
model = pf.GASX(formula="y ~ x1", data=data, ar=1, sc=1, family=pf.Poisson())
x = model.fit()
x.summary()
assert(len(model.predict_is(h=5).values[np.isnan(model.predict_is(h=5).values)]) == 0)
def test_ppc():
"""
Tests PPC value
"""
model = pf.GASX(formula="y ~ x1", data=data, ar=1, sc=1, family=pf.t())
x = model.fit('BBVI', iterations=100)
p_value = model.ppc()
assert (0.0 <= p_value <= 1.0)
def test2_poisson_predict_nans():
"""
Tests that the predictions are not NaNs
"""
model = pf.GASX(formula="y ~ x1 + x2", data=data, ar=1, sc=1, family=pf.Poisson())
x = model.fit()
x.summary()
assert(len(model.predict(h=5, oos_data=data_oos).values[np.isnan(model.predict(h=5,
oos_data=data_oos).values)]) == 0)
def test_predict_nonconstant():
"""
We should not really have predictions that are constant (should be some difference)...
This captures bugs with the predict function not iterating forward
"""
model = pf.GASX(formula="y ~ x1", data=data, ar=1, sc=1, family=pf.Laplace())
x = model.fit()
predictions = model.predict(h=10, oos_data=data_oos, intervals=False)
assert(not np.all(predictions.values==predictions.values[0]))
def test2_sample_model():
"""
Tests sampling function
"""
model = pf.GASX(formula="y ~ x1 + x2", data=data, ar=1, sc=1, family=pf.Laplace())
x = model.fit('BBVI', iterations=100)
sample = model.sample(nsims=100)
assert(sample.shape[0]==100)
assert(sample.shape[1]==len(data)-1)
def test2_skewt_mh():
"""
Tests an GASX model estimated with MEtropolis-Hastings, with multiple predictors, and
tests that the latent variable vector length is correct, and that value are not nan
"""
model = pf.GASX(formula="y ~ x1 + x2", data=data, ar=1, sc=1, family=pf.GASSkewt())
x = model.fit('M-H',nsims=300)
assert(len(model.latent_variables.z_list) == 8)
lvs = np.array([i.value for i in model.latent_variables.z_list])
assert(len(lvs[np.isnan(lvs)]) == 0)
def test_skewt_no_terms(): """ Tests the length of the latent variable vector for an GASX model with no AR or MA terms, and tests that the values are not nan """ model = pf.GASX(formula="y ~ x1", data=data, ar=0, sc=0, family=pf.GASSkewt()) x = model.fit() assert(len(model.latent_variables.z_list) == 5) lvs = np.array([i.value for i in model.latent_variables.z_list]) assert(len(lvs[np.isnan(lvs)]) == 0)
def test_skewt_pml():
"""
Tests an GASX model estimated with PML, and tests that the latent variable
vector length is correct, and that value are not nan
"""
model = pf.GASX(formula="y ~ x1", data=data, ar=1, sc=1, family=pf.GASSkewt())
x = model.fit('PML')
assert(len(model.latent_variables.z_list) == 7)
lvs = np.array([i.value for i in model.latent_variables.z_list])
assert(len(lvs[np.isnan(lvs)]) == 0)
def test_bbvi_mini_batch():
"""
Tests an GASX model estimated with BBVI and that the length of the latent variable
list is correct, and that the estimated latent variables are not nan
"""
model = pf.GASX(formula="y ~ x1", data=data, ar=1, sc=1, family=pf.Laplace())
x = model.fit('BBVI',iterations=500, mini_batch=32)
assert(len(model.latent_variables.z_list) == 5)
lvs = np.array([i.value for i in model.latent_variables.z_list])
assert(len(lvs[np.isnan(lvs)]) == 0)
def test_poisson_couple_terms():
"""
Tests the length of the latent variable vector for an GASX model
with 1 AR and 1 SC term, and tests that the values are not nan
"""
model = pf.GASX(formula="y ~ x1", data=data, ar=1, sc=1, family=pf.Poisson())
x = model.fit()
assert(len(model.latent_variables.z_list) == 4)
lvs = np.array([i.value for i in model.latent_variables.z_list])
assert(len(lvs[np.isnan(lvs)]) == 0)
def test_poisson_laplace():
"""
Tests an GASX model estimated with Laplace approximation, and tests that the latent variable
vector length is correct, and that value are not nan
"""
model = pf.GASX(formula="y ~ x1", data=data, ar=1, sc=1, family=pf.Poisson())
x = model.fit('Laplace')
assert(len(model.latent_variables.z_list) == 4)
lvs = np.array([i.value for i in model.latent_variables.z_list])
assert(len(lvs[np.isnan(lvs)]) == 0)
def test2_laplace():
"""
Tests an GASX model estimated with Laplace, with multiple predictors, and
tests that the latent variable vector length is correct, and that value are not nan
"""
model = pf.GASX(formula="y ~ x1 + x2", data=data, ar=1, sc=1, family=pf.Laplace())
x = model.fit('Laplace')
assert(len(model.latent_variables.z_list) == 6)
lvs = np.array([i.value for i in model.latent_variables.z_list])
assert(len(lvs[np.isnan(lvs)]) == 0)
def test_poisson_mh():
"""
Tests an GASX model estimated with Metropolis-Hastings, and tests that the latent variable
vector length is correct, and that value are not nan
"""
model = pf.GASX(formula="y ~ x1", data=data, ar=1, sc=1, family=pf.Poisson())
x = model.fit('M-H',nsims=300)
assert(len(model.latent_variables.z_list) == 4)
lvs = np.array([i.value for i in model.latent_variables.z_list])
assert(len(lvs[np.isnan(lvs)]) == 0)
def test2_skewt_bbvi():
"""
Tests an GASX model estimated with BBVI, with multiple predictors, and
tests that the latent variable vector length is correct, and that value are not nan
"""
model = pf.GASX(formula="y ~ x1 + x2", data=data, ar=1, sc=1, family=pf.GASSkewt())
x = model.fit('BBVI',iterations=100)
assert(len(model.latent_variables.z_list) == 8)
lvs = np.array([i.value for i in model.latent_variables.z_list])
assert(len(lvs[np.isnan(lvs)]) == 0)
def test2_predict_is_intervals_mh():
"""
Tests prediction intervals are ordered correctly
"""
model = pf.GASX(formula="y ~ x1 + x2", data=data, ar=1, sc=1, family=pf.Poisson())
x = model.fit('M-H', nsims=400)
predictions = model.predict_is(h=10, intervals=True)
assert(np.all(predictions['99% Prediction Interval'].values > predictions['95% Prediction Interval'].values))
assert(np.all(predictions['95% Prediction Interval'].values > predictions['5% Prediction Interval'].values))
assert(np.all(predictions['5% Prediction Interval'].values > predictions['1% Prediction Interval'].values))
def test2_predict_is_intervals_bbvi():
"""
Tests prediction intervals are ordered correctly
"""
model = pf.GASX(formula="y ~ x1 + x2", data=data, ar=1, sc=1, family=pf.Laplace())
x = model.fit('BBVI', iterations=100)
predictions = model.predict_is(h=10, intervals=True)
assert(np.all(predictions['99% Prediction Interval'].values > predictions['95% Prediction Interval'].values))
assert(np.all(predictions['95% Prediction Interval'].values > predictions[model.data_name].values))
assert(np.all(predictions[model.data_name].values > predictions['5% Prediction Interval'].values))
assert(np.all(predictions['5% Prediction Interval'].values > predictions['1% Prediction Interval'].values))
def test2_predict_is_nonconstant():
"""
We should not really have predictions that are constant (should be some difference)...
This captures bugs with the predict function not iterating forward
"""
model = pf.GASX(formula="y ~ x1 + x2",
data=data,
ar=1,
sc=1,
family=pf.Exponential())
assert (not np.all(predictions.values == predictions.values[0]))
def test_predict_is_length():
"""
Tests that the length of the predict IS dataframe is equal to no of steps h
"""
model = pf.GASX(formula="y ~ x1",
data=data,
ar=2,
sc=2,
family=pf.GASNormal())
x = model.fit()
assert (model.predict_is(h=5).shape[0] == 5)
def test2_predict_intervals():
"""
Tests prediction intervals are ordered correctly
"""
model = pf.GASX(formula="y ~ x1 + x2", data=data, ar=1, sc=1, family=pf.Poisson())
model.fit()
predictions = model.predict(h=10, oos_data=data_oos, intervals=True)
assert(np.all(predictions['99% Prediction Interval'].values > predictions['95% Prediction Interval'].values))
assert(np.all(predictions['95% Prediction Interval'].values > predictions['5% Prediction Interval'].values))
assert(np.all(predictions['5% Prediction Interval'].values > predictions['1% Prediction Interval'].values))
def test2_poisson_no_terms():
"""
Tests the length of the latent variable vector for an GASX model
with no AR or SC terms, and two predictors, and tests that the values
are not nan
"""
model = pf.GASX(formula="y ~ x1 + x2", data=data, ar=0, sc=0, family=pf.Poisson())
x = model.fit()
assert(len(model.latent_variables.z_list) == 3)
lvs = np.array([i.value for i in model.latent_variables.z_list])
assert(len(lvs[np.isnan(lvs)]) == 0)
def test2_bbvi_elbo():
"""
Tests that the ELBO increases
"""
model = pf.GASX(formula="y ~ x1 + x2",
data=data,
ar=1,
sc=1,
family=pf.Cauchy())
x = model.fit('BBVI', iterations=500, record_elbo=True, map_start=False)
assert (x.elbo_records[-1] > x.elbo_records[0])
def test_skewt_couple_terms_integ(): """ Tests the length of the latent variable vector for an GASX model with 1 AR and 1 MA term and integrated once, and tests that the values are not nan """ model = pf.GASX(formula="y ~ x1", data=data, ar=1, sc=1, integ=1, family=pf.GASSkewt()) x = model.fit() assert(len(model.latent_variables.z_list) == 7) lvs = np.array([i.value for i in model.latent_variables.z_list]) assert(len(lvs[np.isnan(lvs)]) == 0)
def test2_laplace_predict_is_length():
"""
Tests that the length of the predict IS dataframe is equal to no of steps h
"""
model = pf.GASX(formula="y ~ x1 + x2",
data=data,
ar=1,
sc=1,
family=pf.GASLaplace())
x = model.fit()
assert (model.predict_is(h=5).shape[0] == 5)
def test_predict_intervals_mh():
"""
Tests prediction intervals are ordered correctly
"""
model = pf.GASX(formula="y ~ x1", data=data, ar=1, sc=1, family=pf.Laplace())
x = model.fit('M-H', nsims=400)
predictions = model.predict(h=10, oos_data=data_oos, intervals=True)
assert(np.all(predictions['99% Prediction Interval'].values > predictions['95% Prediction Interval'].values))
assert(np.all(predictions['95% Prediction Interval'].values > predictions[model.data_name].values))
assert(np.all(predictions[model.data_name].values > predictions['5% Prediction Interval'].values))
assert(np.all(predictions['5% Prediction Interval'].values > predictions['1% Prediction Interval'].values))
def test_laplace_predict_nans():
"""
Tests that the predictions are not NaNs
"""
model = pf.GASX(formula="y ~ x1",
data=data,
ar=2,
sc=2,
family=pf.GASLaplace())
x = model.fit()
x.summary()
assert (len(
model.predict(h=5, oos_data=data_oos).values[np.isnan(
model.predict(h=5, oos_data=data_oos).values)]) == 0)
def test_exponential_bbvi():
"""
Tests an GASX model estimated with BBVI, and tests that the latent variable
vector length is correct, and that value are not nan
"""
model = pf.GASX(formula="y ~ x1",
data=data,
ar=1,
sc=1,
family=pf.Exponential())
x = model.fit('BBVI', iterations=100)
assert (len(model.latent_variables.z_list) == 4)
lvs = np.array([i.value for i in model.latent_variables.z_list])
assert (len(lvs[np.isnan(lvs)]) == 0)
def test_bbvi_mini_batch_elbo():
"""
Tests that the ELBO increases
"""
model = pf.GASX(formula="y ~ x1",
data=data,
ar=1,
sc=1,
family=pf.Normal())
x = model.fit('BBVI',
iterations=500,
mini_batch=32,
record_elbo=True,
map_start=False)
assert (x.elbo_records[-1] > x.elbo_records[0])
def test2_laplace_couple_terms():
"""
Tests the length of the latent variable vector for an GASX model
with 1 AR and 1 MA term, and two predictors, and tests that the values
are not nan
"""
model = pf.GASX(formula="y ~ x1 + x2",
data=data,
ar=1,
sc=1,
family=pf.GASLaplace())
x = model.fit()
assert (len(model.latent_variables.z_list) == 6)
lvs = np.array([i.value for i in model.latent_variables.z_list])
assert (len(lvs[np.isnan(lvs)]) == 0)
def test_predict_is_intervals():
"""
Tests prediction intervals are ordered correctly
"""
model = pf.GASX(formula="y ~ x1",
data=data,
ar=1,
sc=1,
family=pf.Exponential())
model.fit()
predictions = model.predict_is(h=10, intervals=True)
assert (np.all(predictions['99% Prediction Interval'].values >
predictions['95% Prediction Interval'].values))
assert (np.all(predictions['95% Prediction Interval'].values >
predictions['5% Prediction Interval'].values))
assert (np.all(predictions['5% Prediction Interval'].values >
predictions['1% Prediction Interval'].values))
