I have a date input 'date_dob' which is '20-Apr-53' I tried converting this to format yyyy-mm-dd using following code:

print pd.to_datetime(date_dob,format = '%d-%b-%y')

which returns '2053-04-20 00:00:00' instead of '1953-04-20 00:00:00'

You need to explicitly check for the year, and then subtract it by 100 years as:

>>> import datetime>>> my_date = '20-Apr-53'
>>> d = datetime.datetime.strptime(my_date, '%d-%b-%y')
>>> d
datetime.datetime(2053, 4, 20, 0, 0)
#                   ^ year = 2053  
>>> if d.year > 2000:
...     d = d.replace(year=d.year-100)  # subtract year by 100
... 
>>> d
datetime.datetime(1953, 4, 20, 0, 0)
#                   ^ year = 1953
>>> d.strftime('%Y-%m-%d') # %Y instead of %y, reason mentioned below
'1953-04-20'
#  ^ all four digit of year

Using %Y instead of %y because:

• %y: displays last two digit of year

• %Y: display all four digit of year