Eliminate Panel for multiple symbols, return MultiIndex DataFrame instead

Author: David Stephens

docs/source/whatsnew/v0.7.0.txt

@@ -42,6 +42,10 @@ Enhancements
 
 Backwards incompatible API changes
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+- When requesting multiple symbols from a DailyReader (ex: google, yahoo, IEX)
+  a MultiIndex DataFrame is now returned.  Previously Panel or dict of DataFrames
+  were returned. (:issue:`297`).
+
 
 .. _whatsnew_070.bug_fixes:
 

pandas_datareader/base.py

@@ -5,8 +5,8 @@
 import requests
 
 import pandas.compat as compat
-from pandas import Panel, DataFrame
-from pandas import read_csv
+from pandas import DataFrame
+from pandas import read_csv, concat
 from pandas.io.common import urlencode
 from pandas.compat import StringIO, bytes_to_str
 
@@ -239,7 +239,9 @@ def _dl_mult_symbols(self, symbols):
                 df_na[:] = np.nan
                 for sym in failed:
                     stocks[sym] = df_na
-            return Panel(stocks).swapaxes('items', 'minor')
+            result = concat(stocks).unstack(level=0)
+            result.columns.names = ['Attributes', 'Symbols']
+            return result
         except AttributeError:
             # cannot construct a panel with just 1D nans indicating no data
             msg = "No data fetched using {0!r}"

pandas_datareader/google/daily.py

@@ -9,9 +9,10 @@
 
 class GoogleDailyReader(_DailyBaseReader):
     """
-    Returns DataFrame/Panel of historical stock prices from symbols, over date
-    range, start to end. To avoid being penalized by Google Finance servers,
-    pauses between downloading 'chunks' of symbols can be specified.
+    Returns DataFrame of historical stock prices from
+    symbols, over date range, start to end. To avoid being penalized by Google
+    Finance servers, pauses between downloading 'chunks' of symbols can be
+    specified.
 
     Parameters
     ----------

pandas_datareader/iex/daily.py

@@ -16,9 +16,10 @@
 class IEXDailyReader(_DailyBaseReader):
 
     """
-    Returns DataFrame/Panel of historical stock prices from symbols, over date
-    range, start to end. To avoid being penalized by Google Finance servers,
-    pauses between downloading 'chunks' of symbols can be specified.
+    Returns DataFrame of historical stock prices
+    from symbols, over date range, start to end. To avoid being penalized by
+    IEX servers, pauses between downloading 'chunks' of symbols can be
+    specified.
 
     Parameters
     ----------
@@ -110,5 +111,7 @@ def _read_lines(self, out):
             df = df.loc[sstart:send]
             result.update({symbol: df})
         if len(result) > 1:
+            result = pd.concat(result).unstack(level=0)
+            result.columns.names = ['Attributes', 'Symbols']
             return result
         return result[self.symbols]

pandas_datareader/stooq.py

@@ -4,8 +4,8 @@
 class StooqDailyReader(_DailyBaseReader):
 
     """
-    Returns DataFrame/Panel of historical stock prices from symbols, over date
-    range, start to end.
+    Returns DataFrame/dict of Dataframes of historical stock prices from
+    symbols, over date range, start to end.
 
     Parameters
     ----------
@@ -18,7 +18,7 @@ class StooqDailyReader(_DailyBaseReader):
         Time, in seconds, to pause between consecutive queries of chunks. If
         single value given for symbol, represents the pause between retries.
     chunksize : int, default 25
-        Number of symbols to download consecutively before intiating pause.
+        Number of symbols to download consecutively before initiating pause.
     session : Session, default None
         requests.sessions.Session instance to be used
 

pandas_datareader/tests/google/test_google.py

@@ -126,8 +126,8 @@ def test_get_multi1(self):
     @skip_on_exception(RemoteDataError)
     def test_get_multi_invalid(self):
         sl = ['AAPL', 'AMZN', 'INVALID']
-        pan = web.get_data_google(sl, '2012')
-        assert 'INVALID' in pan.minor_axis
+        data = web.get_data_google(sl, '2012')
+        assert 'INVALID' in data.columns.levels[1]
 
     def test_get_multi_all_invalid(self):
         sl = ['INVALID', 'INVALID2', 'INVALID3']

pandas_datareader/tests/test_iex_daily.py

@@ -1,5 +1,7 @@
 from datetime import datetime
 
+from pandas import DataFrame, MultiIndex
+
 import pytest
 
 import pandas_datareader.data as web
@@ -43,21 +45,22 @@ def test_single_symbol(self):
     def test_multiple_symbols(self):
         syms = ["AAPL", "MSFT", "TSLA"]
         df = web.DataReader(syms, "iex", self.start, self.end)
-        assert sorted(list(df)) == syms
+        assert sorted(list(df.columns.levels[1])) == syms
         for sym in syms:
-            assert len(df[sym] == 578)
+            assert len(df.xs(sym, level='Symbols', axis=1) == 578)
 
     def test_multiple_symbols_2(self):
         syms = ["AAPL", "MSFT", "TSLA"]
         good_start = datetime(2017, 2, 9)
         good_end = datetime(2017, 5, 24)
         df = web.DataReader(syms, "iex", good_start, good_end)
-        assert isinstance(df, dict)
-        assert len(df) == 3
-        assert sorted(list(df)) == syms
+        assert isinstance(df, DataFrame)
+        assert isinstance(df.columns, MultiIndex)
+        assert len(df.columns.levels[1]) == 3
+        assert sorted(list(df.columns.levels[1])) == syms
 
-        a = df["AAPL"]
-        t = df["TSLA"]
+        a = df.xs("AAPL", axis=1, level='Symbols')
+        t = df.xs("TSLA", axis=1, level='Symbols')
 
         assert len(a) == 73
         assert len(t) == 73

pandas_datareader/tests/yahoo/test_yahoo.py

@@ -171,9 +171,9 @@ def test_get_data_null_as_missing_data(self, adj_pr):
 
     @skip_on_exception(RemoteDataError)
     def test_get_data_multiple_symbols_two_dates(self):
-        pan = web.get_data_yahoo(['GE', 'MSFT', 'INTC'], 'JAN-01-12',
-                                 'JAN-31-12')
-        result = pan.Close.loc['01-18-12'].T
+        data = web.get_data_yahoo(['GE', 'MSFT', 'INTC'], 'JAN-01-12',
+                                  'JAN-31-12')
+        result = data.Close.loc['01-18-12'].T
         assert result.size == 3
 
         # sanity checking
@@ -183,7 +183,7 @@ def test_get_data_multiple_symbols_two_dates(self):
                              [18.58, 28.31, 25.13],
                              [19.03, 28.16, 25.52],
                              [18.81, 28.82, 25.87]])
-        df = pan.Open
+        df = data.Open
         result = df[(df.index >= 'Jan-15-12') & (df.index <= 'Jan-20-12')]
         assert expected.shape == result.shape
 
@@ -279,4 +279,4 @@ def test_yahoo_DataReader_multi(self):
         start = datetime(2010, 1, 1)
         end = datetime(2015, 5, 9)
         result = web.DataReader(['AAPL', 'F'], 'yahoo-actions', start, end)
-        assert isinstance(result, pd.Panel)
+        assert isinstance(result, dict)