I have a dataframe with values similar to below
A10d B10d C10d A B C Strategy
20 10 5 3 5 1 3
The Strategy selects the max of A10d, B10d, C10d and return the value of A,B,C In this case A10d is the largest and Strategy returns A, value of 3
I am not sure how to create this Strategy column properly, can anyone advise please? Thank you very much for your help
I think you need iloc for select first columns by positions and then get columns names by max with idxmax and replace 10d by whitespace for match columns. Last create new column by lookup:
print (df)
A10d B10d C10d A B C
0 20 10 5 3 5 1
1 20 100 5 3 5 1
df1 = df.iloc[:,:3]
print (df1)
A10d B10d C10d
0 20 10 5
1 20 100 5
s = df1.idxmax(axis=1).str.replace('10d','')
print (s)
0 A
1 B
dtype: object
df['Strategy'] = df.lookup(df.index, s)
print (df)
A10d B10d C10d A B C Strategy
0 20 10 5 3 5 1 3
1 20 100 5 3 5 1 5
