def test_order1_logp(self):
data = np.array([0.3, 1, 2, 3, 4])
phi = np.array([0.99])
with Model() as t:
y = AR("y", phi, sigma=1, init_dist=Flat.dist(), shape=len(data))
z = Normal("z", mu=phi * data[:-1], sigma=1, shape=len(data) - 1)
ar_like = t.compile_logp(y)({"y": data})
reg_like = t.compile_logp(z)({"z": data[1:]})
np.testing.assert_allclose(ar_like, reg_like)
with Model() as t_constant:
y = AR(
"y",
np.hstack((0.3, phi)),
sigma=1,
init_dist=Flat.dist(),
shape=len(data),
constant=True,
)
z = Normal("z",
mu=0.3 + phi * data[:-1],
sigma=1,
shape=len(data) - 1)
ar_like = t_constant.compile_logp(y)({"y": data})
reg_like = t_constant.compile_logp(z)({"z": data[1:]})
np.testing.assert_allclose(ar_like, reg_like)
def __init__(self, rho, sigma=None, tau=None, constant=False, init=None, *args, **kwargs):
super().__init__(*args, **kwargs)
tau, sigma = get_tau_sigma(tau=tau, sigma=sigma)
self.sigma = at.as_tensor_variable(sigma)
self.tau = at.as_tensor_variable(tau)
self.mean = at.as_tensor_variable(0.0)
if isinstance(rho, list):
p = len(rho)
else:
try:
shape_ = rho.shape.tag.test_value
except AttributeError:
shape_ = rho.shape
if hasattr(shape_, "size") and shape_.size == 0:
p = 1
else:
p = shape_[0]
if constant:
self.p = p - 1
else:
self.p = p
self.constant = constant
self.rho = rho = at.as_tensor_variable(rho)
self.init = init or Flat.dist()
def __init__(
self, mu=0.0, cov=None, tau=None, chol=None, lower=True, init=None, *args, **kwargs
):
super().__init__(*args, **kwargs)
self.init = init or Flat.dist()
self.innovArgs = (mu, cov, tau, chol, lower)
self.innov = multivariate.MvNormal.dist(*self.innovArgs, shape=self.shape)
self.mean = at.as_tensor_variable(0.0)
def test_order2_logp(self):
data = np.array([0.3, 1, 2, 3, 4])
phi = np.array([0.84, 0.10])
with Model() as t:
y = AR("y", phi, sigma=1, init_dist=Flat.dist(), shape=len(data))
z = Normal("z",
mu=phi[0] * data[1:-1] + phi[1] * data[:-2],
sigma=1,
shape=len(data) - 2)
ar_like = t.compile_logp(y)({"y": data})
reg_like = t.compile_logp(z)({"z": data[2:]})
np.testing.assert_allclose(ar_like, reg_like)
def __init__(self, tau=None, init=None, sigma=None, mu=0.0, *args, **kwargs):
kwargs.setdefault("shape", 1)
super().__init__(*args, **kwargs)
if sum(self.shape) == 0:
raise TypeError("GaussianRandomWalk must be supplied a non-zero shape argument!")
tau, sigma = get_tau_sigma(tau=tau, sigma=sigma)
self.tau = at.as_tensor_variable(tau)
sigma = at.as_tensor_variable(sigma)
self.sigma = sigma
self.mu = at.as_tensor_variable(mu)
self.init = init or Flat.dist()
self.mean = at.as_tensor_variable(0.0)
def test_linear():
lam = -0.78
sig2 = 5e-3
N = 300
dt = 1e-1
sde = lambda x, lam: (lam * x, sig2)
x = floatX(_gen_sde_path(sde, (lam, ), dt, N, 5.0))
z = x + np.random.randn(x.size) * sig2
# build model
with Model() as model:
lamh = Flat("lamh")
xh = EulerMaruyama("xh", dt, sde, (lamh, ), shape=N + 1, initval=x)
Normal("zh", mu=xh, sigma=sig2, observed=z)
# invert
with model:
trace = sample(init="advi+adapt_diag", chains=1)
ppc = sample_posterior_predictive(trace, model=model)
p95 = [2.5, 97.5]
lo, hi = np.percentile(trace[lamh], p95, axis=0)
assert (lo < lam) and (lam < hi)
lo, hi = np.percentile(ppc["zh"], p95, axis=0)
assert ((lo < z) * (z < hi)).mean() > 0.95
