CLN: Remove compat nanmean

Drop compat version of nanmean in favor of standard implementation
bashtage · Jun 28, 2017 · 6cb6d49 · 6cb6d49
1 parent cb9a837
commit 6cb6d49
Show file tree

Hide file tree

Showing 3 changed files with 5 additions and 8 deletions.
diff --git a/statsmodels/compat/numpy.py b/statsmodels/compat/numpy.py
@@ -44,7 +44,6 @@
 import numpy as np
 
 np_matrix_rank = np.linalg.matrix_rank
-nanmean = np.nanmean
 
 if NumpyVersion(np.__version__) >= '1.9.0':
     np_new_unique = np.unique

diff --git a/statsmodels/multivariate/pca.py b/statsmodels/multivariate/pca.py
@@ -9,7 +9,6 @@
 import pandas as pd
 
 from statsmodels.compat.python import range
-from statsmodels.compat.numpy import nanmean
 from statsmodels.tools.sm_exceptions import (ValueWarning,
                                              EstimationWarning)
 
@@ -409,8 +408,8 @@ def _prepare_data(self):
         if np.all(np.isnan(adj_data)):
             return np.empty(adj_data.shape[1]).fill(np.nan)
 
-        self._mu = nanmean(adj_data, axis=0)
-        self._sigma = np.sqrt(nanmean((adj_data - self._mu) ** 2.0, axis=0))
+        self._mu = np.nanmean(adj_data, axis=0)
+        self._sigma = np.sqrt(np.nanmean((adj_data - self._mu) ** 2.0, axis=0))
         if self._standardize:
             data = (adj_data - self._mu) / self._sigma
         elif self._demean:
@@ -489,7 +488,7 @@ def _fill_missing_em(self):
         mask = np.isnan(data)
 
         # 4. Compute mean
-        mu = nanmean(data, 0)
+        mu = np.nanmean(data, 0)
 
         # 5. Replace missing with mean
         projection = np.ones((self._nobs, 1)) * mu

diff --git a/statsmodels/multivariate/tests/test_pca.py b/statsmodels/multivariate/tests/test_pca.py
@@ -19,7 +19,6 @@
 
 from statsmodels.multivariate.pca import PCA
 from statsmodels.multivariate.tests.results.datamlw import data, princomp1, princomp2
-from statsmodels.compat.numpy import nanmean
 
 DECIMAL_5 = .00001
 WIN32 = os.name == 'nt' and sys.maxsize < 2**33
@@ -307,9 +306,9 @@ def test_replace_missing(self):
 
         pc = PCA(x, ncomp=3, missing='fill-em')
         missing = np.isnan(x)
-        mu = nanmean(x, axis=0)
+        mu = np.nanmean(x, axis=0)
         errors = x - mu
-        sigma = np.sqrt(nanmean(errors ** 2, axis=0))
+        sigma = np.sqrt(np.nanmean(errors ** 2, axis=0))
         x_std = errors / sigma
         x_std[missing] = 0.0
         last = x_std[missing]