cumulative returns based on pricing pandas dataframe

Asked Dec 24 '17 at 09:10

Active Dec 24 '17 at 11:16

Viewed 468 times

I want to calculate the cumulative returns based on prices for each stock in a pandas dataframe. the cumulative return is current day price - price at the beginning of time (first row )in the data frame -1. How can we get this in one line ?

My dataframe is like this -

date, stock price1 , stock price2, stock_price3 and so on.



                NSE         ITC
   2012-12-24  5855.750000  169.398026
   2012-12-26  5905.600098  171.433975
   2012-12-27  5870.100098  170.194687
   2012-12-28  5908.350098  170.784225
   2012-12-31  5905.100098  169.250168

i need a generic solution that I can execute and get a result in one line.

edited Dec 24 '17 at 09:42

asked Dec 24 '17 at 09:10

Aditya Lodha

`df['new'] = df.groupby('stock')['price'].cumsum()` Change the colnames according to your dataframe. And if you want to get a detailed answer you should read this https://stackoverflow.com/questions/20109391/how-to-make-good-reproducible-pandas-examples – Bharath M Shetty Dec 24 '17 at 09:12
I dont want to iterate over the columns. Is there any other solution ? – Aditya Lodha Dec 24 '17 at 09:21
We need sample data. Its expensive to create datasets on our own that match the question. – Bharath M Shetty Dec 24 '17 at 09:22
There are many solutions. Unfortunately, none without some sample data. – cs95 Dec 24 '17 at 09:31
provided the dataset – Aditya Lodha Dec 24 '17 at 09:43
For getting daily return I used - daily_return = (dataframe / dataframe.shift(1)) - 1 . can I do a similar thing for cumulative return ? – Aditya Lodha Dec 24 '17 at 09:48
1

Please post your desired data set as well. Are you after `df.diff()` or `df.diff().cumsum()`? – MaxU - stand with Ukraine Dec 24 '17 at 09:50

cumulative returns based on pricing pandas dataframe

0 Answers0