DOC: Continue fixing docstring formatting

Fix more docstrings

docs/themes/statsmodels/static/nature.css_t

@@ -363,7 +363,7 @@ div.searchformwrapper {
 .dropdown:hover .dropdown-content {display: block;}
 
 /* Change the background color of the dropdown button when the dropdown content is shown */
-.dropdown:hover .dropbtn {background-color: #ecf1fb;}
+.dropdown:hover .dropbtn {background-color: #4675f3;}
 
 span.pre {
     font-family: 'Roboto Mono', monospace;

statsmodels/base/elastic_net.py

@@ -110,7 +110,8 @@ def fit_elasticnet(model, method="coord_descent", maxiter=100,
 
     Returns
     -------
-    A results object.
+    Results
+        A results object.
 
     Notes
     -----

statsmodels/base/model.py

@@ -1018,8 +1018,8 @@ def predict(self, exog=None, transform=True, *args, **kwargs):
 
         Returns
         -------
-        prediction : ndarray, pandas.Series or pandas.DataFrame
-            See self.model.predict
+        prediction : {ndarray, Series, DataFrame}
+            See self.model.predict.
 
         Notes
         -----
@@ -1038,7 +1038,6 @@ def predict(self, exog=None, transform=True, *args, **kwargs):
 
         Row indices as in pandas data frames are supported, and added to the
         returned prediction.
-
         """
         import pandas as pd
 
@@ -1263,21 +1262,19 @@ class LikelihoodModelResults(Results):
 
     # by default we use normal distribution
     # can be overwritten by instances or subclasses
-    use_t = False
 
     def __init__(self, model, params, normalized_cov_params=None, scale=1.,
                  **kwargs):
         super(LikelihoodModelResults, self).__init__(model, params)
         self.normalized_cov_params = normalized_cov_params
         self.scale = scale
-
+        self._use_t = False
         # robust covariance
         # We put cov_type in kwargs so subclasses can decide in fit whether to
         # use this generic implementation
         if 'use_t' in kwargs:
             use_t = kwargs['use_t']
-            if use_t is not None:
-                self.use_t = use_t
+            self.use_t = use_t if use_t is not None else False
         if 'cov_type' in kwargs:
             cov_type = kwargs.get('cov_type', 'nonrobust')
             cov_kwds = kwargs.get('cov_kwds', {})
@@ -1318,6 +1315,14 @@ def _get_robustcov_results(self, cov_type='nonrobust', use_self=True,
             # TODO: we should not need use_t in get_robustcov_results
             get_robustcov_results(self, cov_type=cov_type, use_self=True,
                                   use_t=use_t, **cov_kwds)
+    @property
+    def use_t(self):
+        """Flag indicating to use the Student's distribution in inference."""
+        return self._use_t
+
+    @use_t.setter
+    def use_t(self, value):
+        self._use_t = bool(value)
 
     @cached_value
     def llf(self):
@@ -1488,9 +1493,9 @@ def t_test(self, r_matrix, cov_p=None, scale=None, use_t=None):
 
         See Also
         --------
-        tvalues : individual t statistics
-        f_test : for F tests
-        patsy.DesignInfo.linear_constraint
+        tvalues : Indivudals t statistics for the estimated parameters.
+        f_test : Perform an F tests on model parameters.
+        patsy.DesignInfo.linear_constraint : Specify a linear constraint.
 
         Examples
         --------
@@ -1635,10 +1640,10 @@ def f_test(self, r_matrix, cov_p=None, scale=1.0, invcov=None):
 
         See Also
         --------
-        statsmodels.stats.contrast.ContrastResults
-        wald_test
-        t_test
-        patsy.DesignInfo.linear_constraint
+        t_test : Perform a single hypothesis test.
+        wald_test : Perform a Wald-test using a quadratic form.
+        statsmodels.stats.contrast.ContrastResults : Test results.
+        patsy.DesignInfo.linear_constraint : Specify a linear constraint.
 
         Notes
         -----
@@ -1750,10 +1755,10 @@ def wald_test(self, r_matrix, cov_p=None, scale=1.0, invcov=None,
 
         See Also
         --------
-        statsmodels.stats.contrast.ContrastResults
-        f_test
-        t_test
-        patsy.DesignInfo.linear_constraint
+        f_test : Perform an F tests on model parameters.
+        t_test : Perform a single hypothesis test.
+        statsmodels.stats.contrast.ContrastResults : Test results.
+        patsy.DesignInfo.linear_constraint : Specify a linear constraint.
 
         Notes
         -----
@@ -1837,7 +1842,7 @@ def wald_test(self, r_matrix, cov_p=None, scale=1.0, invcov=None,
     def wald_test_terms(self, skip_single=False, extra_constraints=None,
                         combine_terms=None):
         """
-        Compute a sequence of Wald tests for terms over multiple columns
+        Compute a sequence of Wald tests for terms over multiple columns.
 
         This computes joined Wald tests for the hypothesis that all
         coefficients corresponding to a `term` are zero.
@@ -1887,7 +1892,6 @@ def wald_test_terms(self, skip_single=False, extra_constraints=None,
         C(Weight):C(Duration)   0.216694     0.897315972824              2
         Duration               11.187849     0.010752286833              3
         Weight                 30.263368  4.32586407145e-06              4
-
         """
         # lazy import
         from collections import defaultdict
@@ -1973,7 +1977,7 @@ def wald_test_terms(self, skip_single=False, extra_constraints=None,
     def t_test_pairwise(self, term_name, method='hs', alpha=0.05,
                         factor_labels=None):
         """
-        Perform pairwise t_test with multiple testing corrected p-values
+        Perform pairwise t_test with multiple testing corrected p-values.
 
         This uses the formula design_info encoding contrast matrix and should
         work for all encodings of a main effect.
@@ -1986,7 +1990,7 @@ def t_test_pairwise(self, term_name, method='hs', alpha=0.05,
             correspond to the main part of the exog names.
         method : {str, list[str]}
             The multiple testing p-value correction to apply. The default is
-            'hs'. See stats.multipletesting
+            'hs'. See stats.multipletesting.
         alpha : float
             The significance level for multiple testing reject decision.
         factor_labels : {list[str], None}
@@ -2142,7 +2146,8 @@ def load(cls, fname):
         return load_pickle(fname)
 
     def remove_data(self):
-        """remove data arrays, all nobs arrays from result and model
+        """
+        Remove data arrays, all nobs arrays from result and model.
 
         This reduces the size of the instance, so it can be pickled with less
         memory. Currently tested for use with predict from an unpickled

statsmodels/base/wrapper.py

@@ -52,7 +52,7 @@ def __setstate__(self, dict_):
 
     def save(self, fname, remove_data=False):
         """
-        Save a pickle of this instance
+        Save a pickle of this instance.
 
         Parameters
         ----------
@@ -73,6 +73,19 @@ def save(self, fname, remove_data=False):
 
     @classmethod
     def load(cls, fname):
+        """
+        Load a pickled results instance.
+
+        Parameters
+        ----------
+        fname : {str, handle}
+            A string filename or a file handle.
+
+        Returns
+        -------
+        Results
+            The unpickled results instance.
+        """
         from statsmodels.iolib.smpickle import load_pickle
         return load_pickle(fname)
 

statsmodels/discrete/conditional_models.py

@@ -158,10 +158,13 @@ def fit_regularized(self,
             If True, the model is refit using only the variables that
             have non-zero coefficients in the regularized fit.  The
             refitted model is not regularized.
+        **kwargs
+            Additional keyword argument that are used when fitting the model.
 
         Returns
         -------
-        An array of parameter estimates.
+        Results
+            A results instance.
         """
 
         from statsmodels.base.elastic_net import fit_elasticnet

statsmodels/discrete/count_model.py

@@ -100,7 +100,7 @@ def __init__(self, endog, exog, exog_infl=None, offset=None,
 
     def loglike(self, params):
         """
-        Loglikelihood of Generic Zero Inflated model
+        Loglikelihood of Generic Zero Inflated model.
 
         Parameters
         ----------
@@ -118,13 +118,12 @@ def loglike(self, params):
         .. math:: \\ln L=\\sum_{y_{i}=0}\\ln(w_{i}+(1-w_{i})*P_{main\\_model})+
             \\sum_{y_{i}>0}(\\ln(1-w_{i})+L_{main\\_model})
             where P - pdf of main model, L - loglike function of main model.
-
         """
         return np.sum(self.loglikeobs(params))
 
     def loglikeobs(self, params):
         """
-        Loglikelihood for observations of Generic Zero Inflated model
+        Loglikelihood for observations of Generic Zero Inflated model.
 
         Parameters
         ----------
@@ -135,7 +134,7 @@ def loglikeobs(self, params):
         -------
         loglike : ndarray
             The log likelihood for each observation of the model evaluated
-            at `params`. See Notes
+            at `params`. See Notes for definition.
 
         Notes
         --------
@@ -144,7 +143,6 @@ def loglikeobs(self, params):
             where P - pdf of main model, L - loglike function of main model.
 
         for observations :math:`i=1,...,n`
-
         """
         params_infl = params[:self.k_inflate]
         params_main = params[self.k_inflate:]

statsmodels/discrete/discrete_model.py

@@ -224,6 +224,7 @@ def fit_regularized(self, start_params=None, method='l1',
                         qc_verbose=False, **kwargs):
         """
         Fit the model using a regularized maximum likelihood.
+
         The regularization method AND the solver used is determined by the
         argument method.
 
@@ -245,29 +246,36 @@ def fit_regularized(self, start_params=None, method='l1',
             Set to True to print convergence messages.
         fargs : tuple
             Extra arguments passed to the likelihood function, i.e.,
-            loglike(x,*args)
+            loglike(x,*args).
         callback : callable callback(xk)
             Called after each iteration, as callback(xk), where xk is the
             current parameter vector.
         retall : bool
             Set to True to return list of solutions at each iteration.
             Available in Results object's mle_retvals attribute.
         alpha : non-negative scalar or numpy array (same size as parameters)
-            The weight multiplying the l1 penalty term
+            The weight multiplying the l1 penalty term.
         trim_mode : 'auto, 'size', or 'off'
             If not 'off', trim (set to zero) parameters that would have been
             zero if the solver reached the theoretical minimum.
             If 'auto', trim params using the Theory above.
-            If 'size', trim params if they have very small absolute value
+            If 'size', trim params if they have very small absolute value.
         size_trim_tol : float or 'auto' (default = 'auto')
-            For use when trim_mode == 'size'
+            Tolerance used when trim_mode == 'size'.
         auto_trim_tol : float
-            For sue when trim_mode == 'auto'.  Use
+            Tolerance used when trim_mode == 'auto'.
         qc_tol : float
             Print warning and do not allow auto trim when (ii) (above) is
             violated by this much.
         qc_verbose : bool
-            If true, print out a full QC report upon failure
+            If true, print out a full QC report upon failure.
+        **kwargs
+            Additional keyword arguments used when fitting the model.
+
+        Returns
+        -------
+        Results
+            A results instance.
 
         Notes
         -----
@@ -290,7 +298,6 @@ def fit_regularized(self, start_params=None, method='l1',
                     number of iterative refinement steps when solving KKT
                     equations (default: 1).
 
-
         Optimization methodology
 
         With :math:`L` the negative log likelihood, we solve the convex but
@@ -3436,31 +3443,31 @@ def llr_pvalue(self):
         """
         return stats.distributions.chi2.sf(self.llr, self.df_model)
 
-    def set_null_options(self, llnull=None, attach_results=True, **kwds):
-        """set fit options for Null (constant-only) model
+    def set_null_options(self, llnull=None, attach_results=True, **kwargs):
+        """
+        Set the fit options for the Null (constant-only) model.
 
         This resets the cache for related attributes which is potentially
         fragile. This only sets the option, the null model is estimated
         when llnull is accessed, if llnull is not yet in cache.
 
         Parameters
         ----------
-        llnull : None or float
+        llnull : {None, float}
             If llnull is not None, then the value will be directly assigned to
             the cached attribute "llnull".
         attach_results : bool
             Sets an internal flag whether the results instance of the null
             model should be attached. By default without calling this method,
             thenull model results are not attached and only the loglikelihood
             value llnull is stored.
-        kwds : keyword arguments
-            `kwds` are directly used as fit keyword arguments for the null
-            model, overriding any provided defaults.
-
-        Returns
-        -------
-        no returns, modifies attributes of this instance
+        **kwargs
+            Additional keyword arguments used as fit keyword arguments for the
+            null model. The override and model default values.
 
+        Notes
+        -----
+        Modifies attributes of this instance, and so has no return.
         """
         # reset cache, note we need to add here anything that depends on
         # llnullor the null model. If something is missing, then the attribute
@@ -3475,7 +3482,7 @@ def set_null_options(self, llnull=None, attach_results=True, **kwds):
         if llnull is not None:
             self._cache['llnull'] = llnull
         self._attach_nullmodel = attach_results
-        self._optim_kwds_null = kwds
+        self._optim_kwds_null = kwargs
 
     @cache_readonly
     def llnull(self):
@@ -3629,7 +3636,8 @@ def get_margeff(self, at='overall', method='dydx', atexog=None,
 
     def summary(self, yname=None, xname=None, title=None, alpha=.05,
                 yname_list=None):
-        """Summarize the Regression Results
+        """
+        Summarize the Regression Results.
 
         Parameters
         ----------
@@ -3652,7 +3660,7 @@ def summary(self, yname=None, xname=None, title=None, alpha=.05,
 
         See Also
         --------
-        statsmodels.iolib.summary.Summary : class to hold summary results
+        statsmodels.iolib.summary.Summary : Class that hold summary results.
         """
 
         top_left = [('Dep. Variable:', None),
@@ -3699,7 +3707,8 @@ def summary(self, yname=None, xname=None, title=None, alpha=.05,
 
     def summary2(self, yname=None, xname=None, title=None, alpha=.05,
                  float_format="%.4f"):
-        """Experimental function to summarize regression results
+        """
+        Experimental function to summarize regression results.
 
         Parameters
         ----------
@@ -3724,7 +3733,7 @@ def summary2(self, yname=None, xname=None, title=None, alpha=.05,
 
         See Also
         --------
-        statsmodels.iolib.summary2.Summary : class to hold summary results
+        statsmodels.iolib.summary2.Summary : Class that holds summary results.
         """
         from statsmodels.iolib import summary2
         smry = summary2.Summary()