BUG: Fix .to_excel() for MultiIndex containing a NaN value pandas-dev#13511

Author: mpuels

doc/source/whatsnew/v0.19.0.txt

@@ -767,3 +767,5 @@ Bug Fixes
 - Bug where ``pd.read_gbq()`` could throw ``ImportError: No module named discovery`` as a result of a naming conflict with another python package called apiclient  (:issue:`13454`)
 - Bug in ``Index.union`` returns an incorrect result with a named empty index (:issue:`13432`)
 - Bugs in ``Index.difference`` and ``DataFrame.join`` raise in Python3 when using mixed-integer indexes (:issue:`13432`, :issue:`12814`)
+
+- Bug in ``.to_excel()`` when DataFrame contains a MultiIndex which contains a label with a NaN value (:issue:`13511`)

pandas/formats/format.py

@@ -1839,7 +1839,12 @@ def _format_hierarchical_rows(self):
                 for spans, levels, labels in zip(level_lengths,
                                                  self.df.index.levels,
                                                  self.df.index.labels):
-                    values = levels.take(labels)
+
+                    if levels._can_hold_na:
+                        values = levels.take(labels, fill_value=True)
+                    else:
+                        values = levels.take(labels)
+
                     for i in spans:
                         if spans[i] > 1:
                             yield ExcelCell(self.rowcounter + i, gcolidx,

pandas/io/tests/test_excel.py

@@ -1328,6 +1328,20 @@ def test_to_excel_multiindex(self):
                             parse_dates=False)
             tm.assert_frame_equal(frame, df)
 
+    # GH13511
+    def test_to_excel_multiindex_nan_label(self):
+        _skip_if_no_xlrd()
+
+        frame = self.frame
+        frame.A = np.arange(len(frame))
+        frame.iloc[0, 0] = None
+        frame.set_index(['A', 'B'], inplace=True)
+
+        with ensure_clean(self.ext) as path:
+            frame.to_excel(path, merge_cells=self.merge_cells)
+            df = read_excel(path, index_col=[0, 1])
+            tm.assert_frame_equal(frame, df)
+
     # Test for Issue 11328. If column indices are integers, make
     # sure they are handled correctly for either setting of
     # merge_cells