Add column based on previous values with pandas (python)

Question

I want to add a new column named bullTrend to my (ohlcv) dataframe which is based on the previous values of 2 columns:

If current & previous 11 rows close price is higher than ema -> value of bullTrend changes to True
If current & previous 11 rows close price is lower than ema -> value of bullTrend changes to False
First values -> value of bullTrend changes to NaN
Don't include last row

Dataset:

             timestamp    open    high     low   close    volume          ema
    0    1591162860000  9490.0  9489.5  9489.5  9489.5       1.0  9489.500000
    1    1591162920000  9489.5  9490.0  9490.0  9490.0     406.0  9489.751250
    2    1591162980000  9490.0  9490.0  9490.0  9490.0     488.0  9489.834997
    3    1591163040000  9490.0  9497.0  9489.5  9489.5   12798.0  9489.749988
    4    1591163100000  9489.5  9497.0  9489.0  9497.0    1866.0  9491.229134
    ..             ...     ...     ...     ...     ...       ...          ...
    495  1591192560000  9524.5  9524.5  9524.0  9524.5    1727.0  9564.513010
    496  1591192620000  9524.5  9524.5  9523.0  9523.0  179978.0  9564.097058
    497  1591192680000  9523.0  9524.0  9523.0  9524.0     582.0  9563.695321
    498  1591192740000  9524.0  9523.0  9523.0  9523.0       2.0  9563.287617
    499  1591192800000  9523.0  9524.0  9523.0  9524.0    1324.0  9562.894044

Example of the column:

     bullTrend 
0    NaN
1    NaN
2    NaN
3    NaN
4    True
..             ...
495  True
496  True
497  False
498  False
499  False

it seems to be similar as stackoverflow.com/questions/22081878/… ? — linpingta
– linpingta, Commented Jun 3, 2020 at 14:33

pzmn · Accepted Answer · 2020-06-03 15:18:45Z

2

This should work:

def isBull(row):
    idx = row.name
    if idx in idxs:
        return row['eta'] < min([df.at[idx-i, 'close'] for i in range(12)])
    else:
        return np.nan

idxs = df.index[12:-1]
df['bullTrend'] = df.apply(isBull, axis=1)

edited Jun 3, 2020 at 15:18

answered Jun 3, 2020 at 15:11

pzmn

1545 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Add column based on previous values with pandas (python)

1 Answer 1

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related