I am trying to get the indices of the missing date by comparing it to a list of un-missed dates, as the following:

a = pd.DatetimeIndex(["2000", "2001", "2002", "2003",
                      "2004", "2005", "2009", "2010"])
b = pd.DatetimeIndex(["2000", "2001", "2002", "2003",
                      "2004", "2005", "2006", "2007",
                      "2008", "2009", "2010"])
a.reindex(b)

I got the following

(DatetimeIndex(['2000-01-01', '2001-01-01', '2002-01-01', '2003-01-01',
                '2004-01-01', '2005-01-01', '2006-01-01', '2007-01-01',
                '2008-01-01', '2009-01-01', '2010-01-01'],
               dtype='datetime64[ns]', freq=None),
 array([ 0,  1,  2,  3,  4,  5, -1, -1, -1,  6,  7]))

I tried to replace all missing value which is -1 to Nan, by using a.reindex(b, fill_value=np.NAN) but I got the following error TypeError: Index.reindex() got an unexpected keyword argument ‘fill_value’

According the pandas documentation fill_vaue is among the parameters of reindex. Any ideas?

I am trying to get the indices of the missing date by comparing it to a list of un-missed dates, as the following:

a = pd.DatetimeIndex(["2000", "2001", "2002", "2003",
                      "2004", "2005", "2009", "2010"])
b = pd.DatetimeIndex(["2000", "2001", "2002", "2003",
                      "2004", "2005", "2006", "2007",
                      "2008", "2009", "2010"])
a.reindex(b)

I got the following

(DatetimeIndex(['2000-01-01', '2001-01-01', '2002-01-01', '2003-01-01',
                '2004-01-01', '2005-01-01', '2006-01-01', '2007-01-01',
                '2008-01-01', '2009-01-01', '2010-01-01'],
               dtype='datetime64[ns]', freq=None),
 array([ 0,  1,  2,  3,  4,  5, -1, -1, -1,  6,  7]))

I tried to replace all missing value which is -1 to Nan, by using a.reindex(b, fill_value=np.NAN) but I got the following error TypeError: Index.reindex() got an unexpected keyword argument ‘fill_value’

According the pandas documentation fill_vaue is among the parameters of reindex. Any ideas?

• python
• pandas
• datetime

• 2 It is because you do not have a DataFrame, but an Index object. Here are the relevant docs: pandas.pydata.org/docs/reference/api/pandas.Index.reindex.html. – Dr. V Commented Jan 18 at 18:23

2 Answers 2

First of all, you have to do:

newIndex, indexer = a.reindex(b)

reindex returns two things. You need/want to get only the indexer.

So now you can get what you want:

indexerWithNan = np.where(indexer == -1, np.nan, indexer)

Which is:

[ 0.  1.  2.  3.  4.  5. nan nan nan  6.  7.]

Why was your initial code wrong? The reindex() method does not support the fill_value parameter for pandas.Index objects as it does for pandas.Series or pandas.DataFrame.

If you just want to know which values of b are missing from a use Index.isin:

~b.isin(a)

Output:

array([False, False, False, False, False, False,  True,  True,  True,
       False, False])

If you want the missing values:

b[~b.isin(a)]

Output:

DatetimeIndex(['2006-01-01', '2007-01-01', '2008-01-01'], dtype='datetime64[ns]', freq=None)

and for the indices:

np.where(~b.isin(a))[0]

Output: array([6, 7, 8])