def lambda_(self):
df = self._df
rho = self._rhos
if np.isinf(df):
res = (self._rhos == 1).astype(float)
else:
res = 2 * t.cdf(-np.sqrt((df + 1) * (1 - rho)), df=df + 1)
return TailDep(res, res)
def lambda_(self):
return TailDep(0, 2 * 2**(1 / self.params))
def lambda_(self):
res = (self._rhos == 1).astype(float)
return TailDep(res, res)
def lambda_(self):
return TailDep(0, 0)
def lambda_(self): # pragma: no cover
return TailDep(0, 0)
def lambda_(self):
if np.isnan(self._theta):
return TailDep(self._theta, self._theta)
return TailDep(2**(-1 / self._theta) if self._theta > 0 else 0, 0)
