Python pandas fill missing value (NaN) based on condition of another column

Question

I have figured out how to fill the NaN values with the previous cell by using df.fillna(method='ffill') .

However, I am not sure how to base it on a condition that if the country name differs from the country name in its previous cell, then the total case cell value should be 0, otherwise replace NaN with the previous cell's total case value.

Answer 1

Simply using groupby with fillna will give the wanted result. columns here are all the columns you want to apply the missing value logic to.

columns = ['total_cases', 'total_deaths', ...]
df[columns] = df.groupby('location')[columns].fillna(method='ffill').fillna(0)

Note that you need to apply fillna twice, once with forward fill and once with a constant 0 to fill all nan values. This is to make sure that any nans that starts in a new group are filled with 0.