ENH/TST: add pred_table, adjust and add unit tests

statsmodels/examples/ex_ordered_model.py

@@ -24,7 +24,7 @@
 print(np.bincount(y))
 
 mod = OrderedModel(y, x)
-start_params = np.ones(k_vars + 4)
+# start_params = np.ones(k_vars + 4)
 start_params = np.concatenate((np.ones(k_vars), np.arange(4)))
 start_ppf = stats.norm.ppf((np.bincount(y) / len(y)).cumsum())
 start_threshold = np.concatenate((start_ppf[:1],

statsmodels/miscmodels/ordinal_model.py

@@ -118,7 +118,7 @@ def _check_inputs(self, endog, exog):
             exog_name = ([exog.name] if isinstance(exog, pd.Series)
                          else exog.columns.tolist())
             names['xname'] = exog_name
-            exog = np.asarray(exog)
+            # exog = np.asarray(exog)
 
         if isinstance(endog, pd.Series):
             if isinstance(endog.dtypes, CategoricalDtype):
@@ -315,6 +315,22 @@ def fit(self, start_params=None, method='nm', maxiter=500, full_output=1,
 
 
 class OrderedResults(GenericLikelihoodModelResults):
+
+    def pred_table(self):
+        """prediction table
+
+        returns pandas DataFrame
+
+        """
+        # todo: add category labels
+        categories = np.arange(self.model.k_levels)
+        observed = pd.Categorical(self.model.endog,
+                                  categories=categories, ordered=True)
+        predicted = pd.Categorical(self.predict().argmax(1),
+                                   categories=categories, ordered=True)
+        table = pd.crosstab(observed, predicted, margins=True, dropna=False)
+        return table
+
     @Appender(GenericLikelihoodModelResults.summary.__doc__)
     def summary(self, yname=None, xname=None, title=None, alpha=.05):
         names = self.model.names

statsmodels/miscmodels/tests/test_ordinal_model.py

@@ -5,7 +5,7 @@
 import numpy as np
 import scipy.stats as stats
 
-from numpy.testing import assert_allclose
+from numpy.testing import assert_allclose, assert_equal
 from .results.results_ordinal_model import data_store as ds
 from statsmodels.miscmodels.ordinal_model import OrderedModel
 
@@ -70,6 +70,27 @@ def test_unordered(self):
         assert_allclose(res1.model.endog, resf.model.endog, rtol=1e-10)
         assert_allclose(res1.model.exog, resf.model.exog, rtol=1e-10)
 
+    def test_results_other(self):
+
+        res1 = self.res1
+
+        # results
+        if hasattr(self, "pred_table"):
+            table = res1.pred_table()
+            assert_equal(table.values, self.pred_table)
+
+        # smoke test
+        res1.summary()
+
+        # inherited
+        tt = res1.t_test(np.eye(len(res1.params)))
+        assert_allclose(tt.pvalue, res1.pvalues, rtol=1e-13)
+        # TODO: test using string definition of constraints
+
+        pred = res1.predict(exog=res1.model.exog[-5:])
+        fitted = res1.predict()
+        assert_allclose(pred, fitted[-5:], rtol=1e-13)
+
 
 class TestLogitModel(CheckOrdinalModelMixin):
 
@@ -151,6 +172,12 @@ def setup_class(cls):
         cls.resf = resf
         cls.resu = resu
 
+        # regression numbers
+        cls.pred_table = np.array([[202,  18,   0, 220],
+                                   [112,  28,   0, 140],
+                                   [ 27,  13,   0,  40],  # noqa
+                                   [341,  59,   0, 400]], dtype=np.int64)
+
     def test_loglikerelated(self):
 
         res1 = self.res1
@@ -161,8 +188,8 @@ def test_loglikerelated(self):
         score1 = mod.score(res1.params * fact)
         score_obs_numdiff = mod.score_obs(res1.params * fact)
         score_obs_exog = mod.score_obs_(res1.params * fact)
-        assert_allclose(score_obs_numdiff.sum(0), score1, rtol=1e-8)
-        assert_allclose(score_obs_exog.sum(0), score1[:mod.k_vars], rtol=1e-7)
+        assert_allclose(score_obs_numdiff.sum(0), score1, atol=1e-7)
+        assert_allclose(score_obs_exog.sum(0), score1[:mod.k_vars], atol=1e-7)
 
         # null model
         mod_null = OrderedModel(mod.endog, None,