def test_nominal(self):
family = Multinomial(3)
endog, exog, groups = load_data("gee_nominal_1.csv", icept=False)
# Test with independence correlation
v = Independence()
md = GEE(endog, exog, groups, None, family, v)
md.setup_nominal()
mdf1 = md.fit()
# From statsmodels.GEE (not an independent test)
cf1 = np.r_[0.44944752, 0.45569985, -0.92007064, -0.46766728]
se1 = np.r_[0.09801821, 0.07718842, 0.13229421, 0.08544553]
assert_almost_equal(mdf1.params, cf1, decimal=5)
assert_almost_equal(mdf1.standard_errors(), se1, decimal=5)
# Test with global odds ratio dependence
v = GlobalOddsRatio("nominal")
md = GEE(endog, exog, groups, None, family, v)
md.setup_nominal()
mdf2 = md.fit(start_params=mdf1.params)
# From statsmodels.GEE (not an independent test)
cf2 = np.r_[0.45397549, 0.42278345, -0.91997131, -0.50115943]
se2 = np.r_[0.09646057, 0.07405713, 0.1324629, 0.09025019]
assert_almost_equal(mdf2.params, cf2, decimal=5)
assert_almost_equal(mdf2.standard_errors(), se2, decimal=5)
def test_ordinal_pandas(self):
family = Binomial()
endog_orig, exog_orig, groups = load_data("gee_ordinal_1.csv",
icept=False)
data = np.concatenate(
(endog_orig[:, None], exog_orig, groups[:, None]), axis=1)
data = pd.DataFrame(data)
data.columns = ["endog", "x1", "x2", "x3", "x4", "x5", "group"]
# Recode as cumulative indicators
endog, exog, intercepts, nlevel = \
gee_setup_ordinal(data, "endog")
exog1 = np.concatenate((intercepts, exog), axis=1)
groups = exog1[:, -1]
exog1 = exog1[:, 0:-1]
v = GlobalOddsRatio(nlevel, "ordinal")
beta = gee_ordinal_starting_values(endog_orig, exog_orig.shape[1])
md = GEE(endog, exog1, groups, None, family, v)
mdf = md.fit(start_params=beta)
cf = np.r_[1.09238131, 0.02148193, -0.39879146, -0.01855666,
0.02983409, 1.18123172, 0.01845318, -1.10233886]
se = np.r_[0.10878752, 0.10326078, 0.11171241, 0.05488705, 0.05995019,
0.0916574, 0.05951445, 0.08539281]
assert_almost_equal(mdf.params, cf, decimal=2)
assert_almost_equal(mdf.bse, se, decimal=2)
def test_nominal(self):
family = Multinomial(3)
endog, exog, groups = load_data("gee_nominal_1.csv", icept=False)
# Test with independence correlation
v = Independence()
md = NominalGEE(endog, exog, groups, None, family, v)
mdf1 = md.fit()
# From statsmodels.GEE (not an independent test)
cf1 = np.r_[0.44944752, 0.45569985, -0.92007064, -0.46766728]
se1 = np.r_[0.09801821, 0.07718842, 0.13229421, 0.08544553]
assert_almost_equal(mdf1.params, cf1, decimal=5)
assert_almost_equal(mdf1.standard_errors(), se1, decimal=5)
# Test with global odds ratio dependence
v = GlobalOddsRatio("nominal")
md = NominalGEE(endog, exog, groups, None, family, v)
mdf2 = md.fit(start_params=mdf1.params)
# From statsmodels.GEE (not an independent test)
cf2 = np.r_[0.45448248, 0.41945568, -0.92008924, -0.50485758]
se2 = np.r_[0.09632274, 0.07433944, 0.13264646, 0.0911768]
assert_almost_equal(mdf2.params, cf2, decimal=5)
assert_almost_equal(mdf2.standard_errors(), se2, decimal=5)
def gendat_nominal():
ns = nominal_simulator()
# The last component of params must be identically zero
ns.params = [np.r_[0., 1], np.r_[-1., 0], np.r_[0., 0]]
ns.ngroups = 200
ns.dparams = [
1.,
]
ns.simulate()
data = np.concatenate((ns.endog[:, None], ns.exog, ns.group[:, None]),
axis=1)
ns.endog_ex, ns.exog_ex, ns.exog_ne, ns.nlevel = \
gee_setup_nominal(data, 0, [3,])
ns.group_ex = ns.exog_ne[:, 0]
va = GlobalOddsRatio(3, "nominal")
lhs = np.array([
[0., 1., 1, 0],
])
rhs = np.r_[0., ]
return ns, va, Multinomial(3), (lhs, rhs)
def gendat_ordinal():
os = ordinal_simulator()
os.params = np.r_[0., 1]
os.ngroups = 200
os.thresholds = [1, 0, -1]
os.dparams = [
1.,
]
os.simulate()
data = np.concatenate((os.endog[:, None], os.exog, os.group[:, None]),
axis=1)
os.endog_ex, os.exog_ex, os.intercepts, os.nthresh = \
gee_setup_ordinal(data, 0)
os.group_ex = os.exog_ex[:, -1]
os.exog_ex = os.exog_ex[:, 0:-1]
os.exog_ex = np.concatenate((os.intercepts, os.exog_ex), axis=1)
va = GlobalOddsRatio(4, "ordinal")
lhs = np.array([[0., 0., 0, 1., 0.], [0., 0, 0, 0, 1]])
rhs = np.r_[0., 1]
return os, va, Binomial(), (lhs, rhs)
def test_nominal(self):
family = Multinomial(3)
endog, exog, groups = load_data("gee_nominal_1.csv", icept=False)
# Test with independence correlation
va = Independence()
mod1 = NominalGEE(endog, exog, groups, None, family, va)
rslt1 = mod1.fit()
# Regression test
cf1 = np.r_[0.44944752, 0.45569985, -0.92007064, -0.46766728]
se1 = np.r_[0.09801821, 0.07718842, 0.13229421, 0.08544553]
assert_almost_equal(rslt1.params, cf1, decimal=5)
assert_almost_equal(rslt1.standard_errors(), se1, decimal=5)
# Test with global odds ratio dependence
va = GlobalOddsRatio("nominal")
mod2 = NominalGEE(endog, exog, groups, None, family, va)
rslt2 = mod2.fit(start_params=rslt1.params)
# Regression test
cf2 = np.r_[0.45448248, 0.41945568, -0.92008924, -0.50485758]
se2 = np.r_[0.09632274, 0.07433944, 0.13264646, 0.0911768]
assert_almost_equal(rslt2.params, cf2, decimal=5)
assert_almost_equal(rslt2.standard_errors(), se2, decimal=5)
# Make sure we get the correct results type
assert_equal(type(rslt1), NominalGEEResultsWrapper)
assert_equal(type(rslt1._results), NominalGEEResults)
def setup_class(cls):
family = Binomial()
endog, exog, groups = load_data("gee_ordinal_1.csv", icept=False)
va = GlobalOddsRatio("ordinal")
cls.mod = OrdinalGEE(endog, exog, groups, None, family, va)
cls.start_params = np.array([
1.09250002, 0.0217443, -0.39851092, -0.01812116, 0.03023969,
1.18258516, 0.01803453, -1.10203381
])
def test_wrapper(self):
endog, exog, groups = load_data("gee_ordinal_1.csv", icept=False)
endog = pd.Series(endog, name='yendog')
exog = pd.DataFrame(exog)
groups = pd.Series(groups, name='the_group')
family = Binomial()
va = GlobalOddsRatio("ordinal")
mod = OrdinalGEE(endog, exog, groups, None, family, va)
rslt2 = mod.fit()
check_wrapper(rslt2)
def test_ordinal(self):
family = Binomial()
endog, exog, groups = load_data("gee_ordinal_1.csv", icept=False)
v = GlobalOddsRatio("ordinal")
md = GEE(endog, exog, groups, None, family, v)
md.setup_ordinal()
mdf = md.fit()
cf = np.r_[1.09238131, 0.02148193, -0.39879146, -0.01855666,
0.02983409, 1.18123172, 0.01845318, -1.10233886]
se = np.r_[0.10878752, 0.10326078, 0.11171241, 0.05488705, 0.05995019,
0.0916574, 0.05951445, 0.08539281]
assert_almost_equal(mdf.params, cf, decimal=5)
assert_almost_equal(mdf.bse, se, decimal=5)
def test_ordinal(self):
family = Binomial()
endog, exog, groups = load_data("gee_ordinal_1.csv", icept=False)
v = GlobalOddsRatio("ordinal")
md = OrdinalGEE(endog, exog, groups, None, family, v)
mdf = md.fit()
cf = np.r_[1.09250002, 0.0217443, -0.39851092, -0.01812116, 0.03023969,
1.18258516, 0.01803453, -1.10203381]
se = np.r_[0.10883461, 0.10330197, 0.11177088, 0.05486569, 0.05997153,
0.09168148, 0.05953324, 0.0853862]
assert_almost_equal(mdf.params, cf, decimal=5)
assert_almost_equal(mdf.bse, se, decimal=5)
def test_ordinal(self):
family = Binomial()
endog, exog, groups = load_data("gee_ordinal_1.csv", icept=False)
va = GlobalOddsRatio("ordinal")
mod = OrdinalGEE(endog, exog, groups, None, family, va)
rslt = mod.fit()
# Regression test
cf = np.r_[1.09250002, 0.0217443, -0.39851092, -0.01812116, 0.03023969,
1.18258516, 0.01803453, -1.10203381]
assert_almost_equal(rslt.params, cf, decimal=5)
# Regression test
se = np.r_[0.10883461, 0.10330197, 0.11177088, 0.05486569, 0.05997153,
0.09168148, 0.05953324, 0.0853862]
assert_almost_equal(rslt.bse, se, decimal=5)
# Check that we get the correct results type
assert_equal(type(rslt), OrdinalGEEResultsWrapper)
assert_equal(type(rslt._results), OrdinalGEEResults)