GEE test failure on Debian #1260

Closed
josef-pkt opened this Issue Dec 20, 2013 · 2 comments

Projects

None yet

1 participant

@josef-pkt
Member

http://nipy.bic.berkeley.edu/builders/statsmodels-py2.x-sid-sparc/builds/58/steps/shell_4/logs/stdio

13 times the same error

======================================================================
ERROR: Gaussian GEE with independence correlation should agree
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/buildslave/nd-bb-slave-sparc-sid/statsmodels-py2_x-sid-sparc/build/venv/local/lib/python2.7/site-packages/nose/case.py", line 197, in runTest
    self.test(*self.arg)
  File "/home/buildslave/nd-bb-slave-sparc-sid/statsmodels-py2_x-sid-sparc/build/venv/local/lib/python2.7/site-packages/statsmodels-0.6.0-py2.7-linux-sparc64.egg/statsmodels/genmod/tests/test_gee.py", line 526, in test_compare_OLS
    covstruct=vs)
  File "/home/buildslave/nd-bb-slave-sparc-sid/statsmodels-py2_x-sid-sparc/build/venv/local/lib/python2.7/site-packages/statsmodels-0.6.0-py2.7-linux-sparc64.egg/statsmodels/base/model.py", line 103, in from_formula
    mod = cls(endog, exog, *args, **kwargs)
  File "/home/buildslave/nd-bb-slave-sparc-sid/statsmodels-py2_x-sid-sparc/build/venv/local/lib/python2.7/site-packages/statsmodels-0.6.0-py2.7-linux-sparc64.egg/statsmodels/genmod/generalized_estimating_equations.py", line 292, in __init__
    self.endog_li = self._cluster_list(self.endog)
  File "/home/buildslave/nd-bb-slave-sparc-sid/statsmodels-py2_x-sid-sparc/build/venv/local/lib/python2.7/site-packages/statsmodels-0.6.0-py2.7-linux-sparc64.egg/statsmodels/genmod/generalized_estimating_equations.py", line 362, in _cluster_list
    for k in self.group_labels]
IndexError: too many indices

There is also a precision issue with a test for gmm.IV2SLS

@josef-pkt
Member

fixed, doesn't show up anymore

@josef-pkt josef-pkt closed this Apr 20, 2014
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment