apply a recursive function to a pandas dataframe column

Question

I want to generate a new column based on the function of another column and the row before the current row.

i.e.

NewColumn_{n} = OldColumn_{n} * 0.6 + NewColumn_{n-1} * 0.4

Is it possible I could do this without using iterrows?

I saw this post: How to constuct a column of data frame recursively with pandas-python?

But I'm wondering if there is another function/package or tricks within Pandas?

Something like purrr in R?

Answer 1

You can write a custom func here and use df.apply

def func(s): # s is `Series`
    newColumn = [0]
    for i, val in enumerate(s):
        newColumn.append(val*0.6 + newColumn[i]*0.4)
    return newColumn[1:]

df['new'] = df[['old']].apply(func) #Don't use df['old'] that would call
                                    #Series.apply which applies func element-wise

   old    new
0    1  0.600
1    2  1.440
2    3  2.376