Pandas dataframe, do calulations on any column in rolling window

user_3512126 0
Viewed 268

Maybe pandas.DataFrame.rolling is not the best way to do this, please tell me if there is a better approach.

What I want is to have rolling window on df and have all columns from df available in the window to do various calculations.

I believe the code below is very close to my goals, but I struggle understanding the index issue as stated in the code.

At first x.index = RangeIndex(start=0, stop=2, step=1), and tmp_df correctly selects the first and second row in df (index 0 and 1). For the last x.index = RangeIndex(start=4, stop=6, step=1) it seems like iloc tries to select index 6 in df which is out of range (df has index 0 to 5).

What am I missing?

Thank you in advance for any advice.

import numpy as np
import pandas as pd

df = pd.DataFrame({'open': [7, 5, 10, 11,6,12],
                   'close': [6, 6, 11, 10,7,10],
                   'positive': [0, 1, 1, 0,1,0]},
                 )

def do_calculations_on_any_df_column_in_window(x,df):
    print("index:",x.index)
    tmp_df = df.iloc[x.index] # raises "ValueError: cannot set using a slice indexer with a different length than the value" when x.index = RangeIndex(start=4, stop=6, step=1) as df index goes from 0 to 5 only

    # do calulations on any column in tmp_df, get result
    result = 1 #dummyresult

    return result

intervals = range(2, 10)
for i in intervals:
    df['result_' + str(i)] = np.nan
    res = df.rolling(i).apply(do_calculations_on_any_df_column_in_window, args=(df,), raw=False)
    df['result_' + str(i)][1:] = res

print(df)
rolling-computation valueerror pandas slice python
1 Answers

Try this function:

def calculate_on_rolling_window(df, win, col_names):
    #final_df = pd.DataFrame() # stores the complete results
    # calculate sd and mean for each tag
    for i in range(len(col_names)):
      current_column = col_names[i]
      df[current_column + '_mean_' +str(win)] = (df[current_column].rolling(window=win).mean())
      df[current_column + '_min_' +str(win)] = (df[current_column].rolling(window=win).min())
      df[current_column + '_max_' +str(win)] = (df[current_column].rolling(window=win).max())
    df = df.fillna(0)
    return(df)

You get this results 

col_names = df.columns
df_extended = calculate_on_rolling_window(df,2,col_names)
df_extended.head()

open    close   positive    open_mean_2     open_min_2  open_max_2  close_mean_2    close_min_2     close_max_2     positive_mean_2     positive_min_2  positive_max_2
0     7       6       0           0.0              0.0          0.0        0.0             0.0            0.0             0.0                   0.0           0.0
1     5       6       1           6.0              5.0          7.0        6.0             6.0            6.0             0.5                   0.0           1.0
2     10      11      1           7.5              5.0          10.0       8.5             6.0            11.0            1.0                   1.0           1.0
3     11      10      0           10.5             10.0         11.0       10.5            10.0           11.0            0.5                   0.0           1.0
4     6       7       1           8.5              6.0          11.0       8.5             7.0            10.0            0.5                   0.0           1.0
user_27932280
Related