Home / Questions / 543578
Accepted
NoahVerner
Asked: 2022-08-30 01:03:49 +0800 CST 2022-08-30 01:03:49 +0800 CST

How to correctly add the last value of the index of a df to another empty df? Related to Pandas

  • 772

Let's say I have the following empty df called df_pnl, after calling it it produces the following result:

Empty DataFrame
Columns: [PnL, Type of Trade, ROI, Cumulative ROI, Cumulative Investment]
Index: []

If you run df_pnl.index[-1:], it produces the following output:

Index([], dtype='object', name='End Date')

Now, there is another df called df_chosen_trading_pairthat shows the following data as output when called:

                         Close Price 
End Date                             
2022-06-25 02:29:59.999        11.21
2022-06-25 02:59:59.999        11.19
2022-06-25 03:29:59.999        11.25
2022-06-25 03:59:59.999        11.37
2022-06-25 04:29:59.999        11.34
                             ...    
2022-06-27 04:59:59.999        10.96
2022-06-27 05:29:59.999        10.80
2022-06-27 05:59:59.999        10.82
2022-06-27 06:29:59.999        10.84
2022-06-27 06:59:59.999        10.88

[106 rows x 1 columns]

If you run df_chosen_trading_pair.index[-1], it produces the following output:

'2022-06-27 06:59:59.999'

If you run type(df_chosen_trading_pair.index[-1]), it produces the following output:

str

I need to add the value of df_chosen_trading_pair.index[-1]a df_pnl.index[-1:]to get the following result when calling df_pnl:

                         PnL   Type of Trade  ROI  Cumulative ROI  Cumulative Investment
End Date                             
2022-06-27 06:59:59.999  NaN             NaN  NaN             NaN                    NaN

I tried the following lines and none of them worked:

In[124]: df_pnl.index[0:] = df_chosen_trading_pair.index[-1]

TypeError: Index does not support mutable operations

In[125]: df_pnl["End Date"][-1] = df_chosen_trading_pair.index[-1]

KeyError: 'End Date'

Can you help me, please?

python-3.x

2 Answers

  1. Good day,

    To do so, the dataframevoid must have the columns you need

    Then you can use pandas.DataFrame.reindexto add the index of the otherdataframe

    With a test file "sample.csv" with the data you put in your question, create the following example:

    import pandas as pd

df_e = pd.DataFrame(columns=['End Date', 'PnL', 'Type of Trade', 'ROI', 'Cumulative ROI', 'Cumulative Investment'])
df_e.set_index('End Date')

df = pd.read_csv('sample.csv', index_col = 'End Date')

df_e = df_e.reindex([df.index[-1]])

    Printing df_ewe get:

    
End Date                PnL Type of Trade   ROI Cumulative ROI  Cumulative Investment
2022-06-27 06:59:59.999 NaN NaN             NaN NaN             NaN        NaN
    • 2
  2. Best Answer

    Oops, I think I accidentally found my solution:

    In[167]: df_pnl.loc[df_chosen_trading_pair.index[-1], :] = None

    Generate:

                             PnL   Type of Trade  ROI  Cumulative ROI  Cumulative Investment
End Date                             
2022-06-27 06:59:59.999  NaN             NaN  NaN             NaN                    NaN

    Also, I realized that if for example I want to add a new row using the penultimate index value of df_chosen_trading_pair, I simply have to change the value inside the array, like this:

    In[167]: df_pnl.loc[df_chosen_trading_pair.index[-2], :] = None

    Generate:

                             PnL   Type of Trade  ROI  Cumulative ROI  Cumulative Investment
End Date                             
2022-06-27 06:59:59.999  NaN             NaN  NaN             NaN                    NaN

2022-06-27 06:29:59.999  NaN             NaN  NaN             NaN                    NaN
    • 0