It is not clear if this should raise on construction or not http://stackoverflow.com/questions/17338148/pandas-duplicate-datetimeindex-entries-lead-to-odd-exception?noredirect=1#comment25166923_17338148

In [3]: from collections import OrderedDict

In [4]: import datetime as dt

In [5]: a = [1371215933513120, 1371215933513121, 1371215933513122, 1371215933513122]

In [6]: b = [1,2,3,4]

In [7]: df = pd.DataFrame({'a':a, 'b':b, 'c':[dt.datetime.fromtimestamp(t/1000000.) for t in a]})

In [8]: df.index=pd.DatetimeIndex(df['c'])

In [9]: d = OrderedDict()

In [10]: d['x'] = df

In [11]: p = pd.Panel(d)

In [12]: for y in p.major_axis:
   ....:         print y
   ....:         print p.major_xs(y)
   ....:     
2013-06-14 09:18:53.513120
                            x
a            1371215933513120
b                           1
c  2013-06-14 09:18:53.513120
2013-06-14 09:18:53.513121
                            x
a            1371215933513121
b                           2
c  2013-06-14 09:18:53.513121
2013-06-14 09:18:53.513122
ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()