Slicing with a logical (boolean) expression a Pandas Dataframe

Question

I am getting an exception as I try to slice with a logical expression my Pandas dataframe.

My data have the following form:

df
    GDP_norm    SP500_         


        

Answer 1

You are suffering from the effects of chained comparisons. What's happening is the expression df['GDP_norm'] >=3.5 & df['GDP_norm'] <= 4.5 is evaluated as something like:

df['GDP_norm'] >= (3.5 & df['GDP_norm']) <= 4.5

Of course, this fails since float cannot be compared with bool, as described in your error message. Instead, use parentheses to isolate each Boolean mask and assign to variables:

m1 = (df['GDP_norm'] >= 3.5) & (df['GDP_norm'] <= 4.5)
m2 = df['SP500_Index_deflated_norm'] > 3

m3 = (df['GDP_norm'] >= 4.0) & (df['GDP_norm'] <= 5.0)
m4 = df['SP500_Index_deflated_norm'] < 3.5

res = df[(m1 & m2) | (m3 & m4)]