Pandas Datetime: Calculate Number of Weeks Between Dates in Two Columns

Question

Let's say I have a dataframe with two columns that contain dates, and I want to create a new columns whose value is the number of months between those dates.

>df

Index   Date1         Date2
1       2012/03/07    2013/03/16
2       2012/12/05    2012/12/25
3       2010/06/30    2013/05/19
4       2002/11/02    2011.06.08


df["Date1"]= pd.to_datetime(df["Date1"])
df["Date2"]= pd.to_datetime(df["Date2"])

Date1 will always be before date2. My current method of doing this requires about 10 steps, and I'm pretty sure there's an easier way to do this. Thoughts?

Answer 1

see this link: http://pandas.pydata.org/pandas-docs/dev/timeseries.html#time-deltas

(df['Date2']-df['Date1']).apply(lambda x: x/np.timedelta64(1,'M'))

for numpy >=1.7 (see the link if you are using 1.6.1)

I am not sure what it will do with the fraction. (usually I would divide by np.timedelta64(1,'D') then divide by say 30 to make a fractional num of months (as a float)

Answer 2

I'm not sure how to do it in python but the steps I would do:

• Convert dates into number of days since the epoch
• Subtract date1 from date2
• Divide by 7