CODEWITHSUNDEEP
python-3.xpandasdataframe
Antoine
Antoine

Reputation: 587

Select the maximum/minimum from n previous rows in a DataFrame

I'm using pandas in Python and I have an issue to select some data. I have DataFrame with float values, and I would like to create a column which contains the maximum (or minimum) of the n previous rows of a column, and set to 0 for the n first rows, here's an example of the result I would like to have:

df_test = pd.DataFrame({'a':[2,7,2,0,-1, 19, -52, 2]})
df_test['result_i_want_with_n=3'] = [0, 0, 0, 7, 7, 2, 19, 19]
print(df_test)
    a   result_i_want_with_n=3
0   2   0
1   7   0
2   2   0
3   0   7
4   -1  7
5   19  2
6   -52 19
7   2   19

I managed to get this result using a while, but I would like to program it in a more "pandas" way to gain computation speed.

Thanks

Upvotes: 8

Views: 7784

Answers (1)

P.Tillmann
P.Tillmann

Reputation: 2110

Rolling is your friend here. You need to shift by one row in order to get your exact result, otherwise your first value will be in the third row.

df_test.rolling(window=3).max().shift(1).fillna(0)

0     0.0
1     0.0
2     0.0
3     7.0
4     7.0
5     2.0
6    19.0
7    19.0

Upvotes: 11

Related Questions