dimanche 3 juillet 2016

How to aggregate intraday time stamps and create a new column in pandas dataframe Python

I've tried this problem for hours and has no solution..

Below is intraday 5 min forex data - df. It is recorded every 5 mins every day. I have excluded weekends data since market is close on weekends.Here a weekend is defined as from Friday 5:00 pm - Sunday 5:00 pm.

                   Time             OPEN      CLOSE       
    216    2014-01-01 18:05:00     0.891975  0.892185    
    217    2014-01-01 18:10:00     0.892075  0.892090 
    ...
    210238 2015-12-31 23:55:00     1.000390  1.000390   
    210239 2016-01-01 00:00:00     1.000390  1.000390 

A day is defined as from 11:00 am - 11:00 am. So when I say 2014-01-02, I am aggregating data from 2014-01-01 11:00 am to 2014-01-02 11:00 am. 2014-01-06(monday) will include data from two intervals: 1. 2014-01-03(friday) 11:00 - 17:00 2. 2014-01-05(sunday) 17:00 - 2014-01-06 11:00

I want to create a new column 'Date' to define the new "day" so by reading df.Time, the column will record which day it belongs to.

how would you approach this?

                   Time             OPEN      CLOSE   Date      
        2014-01-03 14:05:00     0.891975  0.892185    2014-01-06
        2014-01-05 17:00:00     0.892075  0.892090    2014-01-06
        2014-01-06 11:00:00     0.892075  0.892090    2014-01-06
    ...
        2015-12-31 23:55:00     1.000390  1.000390    2016-01-01
        2016-01-01 00:00:00     1.000390  1.000390    2016-01-01

Aucun commentaire:

Enregistrer un commentaire