def test_fit_alpha_only(self):
alpha = 0.7
np.random.seed(345)
T = 200
l = 0.2
y = [0] * T
for t in range(0, T):
d = np.random.normal()
y[t] = l + d
l = l + alpha * d
c = Components(use_arma_errors=False)
p = ModelParams(c, alpha=0.09) # default starting value for alpha
optimizer = ParamsOptimizer(self.context)
optimizer.optimize(y, p)
fitted_model = optimizer.optimal_model()
resid = fitted_model.resid
# AGAIN why x0 estimation is so shitty, l is far from being 0.2
# Residuals should form a normal distribution
_, pvalue = stats.normaltest(resid)
assert 0.05 < pvalue # large p-value, we can not reject null hypothesis of normal distribution
# Mean of residuals should be close to 0
_, pvalue = stats.ttest_1samp(resid, popmean=0.0)
assert 0.05 < pvalue # large p-value we can not reject null hypothesis that mean is 0
# We expect 95% of residuals to lie within [-2,2] interval
assert len(resid[np.where(np.abs(resid) < 2)]) / len(resid) > 0.90
def test_calculate_seed_x0(self, components, params, expected):
y = [2.0, 4.0, 2.0]
c = Components(**components)
p = ModelParams(c, **params)
m = ParamsOptimizer(self.context)
x0 = m._calculate_seed_x0(y, p)
assert np.allclose(expected, x0)
def test_with_vector_values(self):
c = Components(use_trend=True, use_damped_trend=True,
use_arma_errors=True, use_box_cox=True, seasonal_periods=[7, 30], p=1, q=2)
p = ModelParams(c, alpha=0.1, box_cox_lambda=1.4, beta=0.2, phi=0.3, gamma_params=[0.4, 0.5],
ar_coefs=[-0.3], ma_coefs=[-0.4, 0.6])
v = p.to_vector()
assert np.array_equal([0.1, 1.4, 0.2, 0.3, 0.4, 0.5, -0.3, -0.4, 0.6], v)
p = ModelParams(c, alpha=0.9, box_cox_lambda=1.4)
p = p.with_vector_values(v)
assert p.alpha == 0.1
assert p.box_cox_lambda == 1.4
assert p.beta == 0.2
assert p.phi == 0.3
assert np.array_equal([0.4, 0.5], p.gamma_params)
assert np.array_equal([-0.3], p.ar_coefs)
assert np.array_equal([-0.4, 0.6], p.ma_coefs)
def test_create_x0_of_zeroes(self, components, params, expected):
c = Components(**components)
p = ModelParams(c, **params)
x0 = p._create_x0_of_zeroes()
assert np.array_equal(expected, x0)
def test_to_vector(self, components, params, expected):
c = Components(**components)
p = ModelParams(c, **params)
vector = p.to_vector()
assert np.array_equal(expected, vector)
def test_gamma(self, components, params, expected_gamma):
c = Components(**components)
p = ModelParams(c, **params)
assert np.array_equal(expected_gamma, p.gamma_params)
def test_make_vector(self, components, params, expected_w, expected_g):
m = MatrixBuilder(ModelParams(Components(**components), **params))
assert np.array_equal(expected_w, m.make_w_vector())
assert np.array_equal(expected_g, m.make_g_vector())
def test_make_F_matrix(self, components, params, expected_matrix):
m = MatrixBuilder(ModelParams(Components(**components), **params))
assert np.array_equal(expected_matrix, m.make_F_matrix())