def cov(self):
"""Estimated covariance"""
x = self._x
nobs = x.shape[0]
xpxi = inv(x.T @ x / nobs)
eps = self.eps
xe = x * eps
if self._clusters.ndim == 1:
xeex = _cov_cluster(xe, self._clusters)
if self._group_debias:
xeex *= self._calc_group_debias(self._clusters)
else:
clusters0 = self._clusters[:, 0]
clusters1 = self._clusters[:, 1]
xeex0 = _cov_cluster(xe, clusters0)
xeex1 = _cov_cluster(xe, clusters1)
sort_keys = np.lexsort(self._clusters.T)
locs = np.arange(self._clusters.shape[0])
lex_sorted = self._clusters[sort_keys]
sorted_locs = locs[sort_keys]
diff = np.any(lex_sorted[1:] != lex_sorted[:-1], 1)
clusters01 = np.cumsum(np.r_[0, diff])
resort_locs = np.argsort(sorted_locs)
clusters01 = clusters01[resort_locs]
xeex01 = _cov_cluster(xe, clusters01)
if self._group_debias:
xeex0 *= self._calc_group_debias(clusters0)
xeex1 *= self._calc_group_debias(clusters1)
xeex01 *= self._calc_group_debias(clusters01)
xeex = xeex0 + xeex1 - xeex01
xeex *= self._scale
out = (xpxi @ xeex @ xpxi) / nobs
return (out + out.T) / 2
def weight_matrix(self, x: NDArray, z: NDArray, eps: NDArray) -> NDArray:
"""
Parameters
----------
x : ndarray
Model regressors (exog and endog), (nobs by nvar)
z : ndarray
Model instruments (exog and instruments), (nobs by ninstr)
eps : ndarray
Model errors (nobs by 1)
Returns
-------
ndarray
Covariance of GMM moment conditions.
"""
nobs, nvar = x.shape
ze = z * eps
mu = ze.mean(axis=0) if self._center else 0
ze -= mu
clusters = self._clusters
if clusters.shape[0] != nobs:
raise ValueError("clusters has the wrong nobs. Expected {0}, "
"got {1}".format(nobs, clusters.shape[0]))
clusters = asarray(clusters).copy().squeeze()
s = _cov_cluster(ze, clusters)
if self._debiased:
num_clusters = len(unique(clusters))
scale = (nobs - 1) / (nobs - nvar) * num_clusters / (num_clusters -
1)
s *= scale
return s
