Question

I have a MultiIndex DataFrame that contains these values:

                      AAPL
           minor          
2007-09-14 OC     0.024436
2007-09-15 CC     0.030293
           CO     0.017518
           OC     0.024688
           OO     0.031835

# to_dict():

{'AAPL': {(<Timestamp: 2007-09-14 00:00:00>, 'OC'): 0.024436265475779286,
  (<Timestamp: 2007-09-15 00:00:00>, 'CC'): 0.030293017084353703,
  (<Timestamp: 2007-09-15 00:00:00>, 'CO'): 0.017518449703066673,
  (<Timestamp: 2007-09-15 00:00:00>, 'OC'): 0.024688182799779634,
  (<Timestamp: 2007-09-15 00:00:00>, 'OO'): 0.031834725061579666}}

—

and a Series that contains these values:

CC    15.874508
CO    18.590320
OC    30.503468
OO    15.874508

# to_dict():

{'CC': 15.874507866387544,
 'CO': 18.590320061795602,
 'OC': 30.503467646507644,
 'OO': 15.874507866387544}

I’d like to multiply all of the minor index CC values by the CC value in the Series, and the same with the other values. I saw another question on here that gave me the .mul method, but when I try that, even with the level=’minor’, it tells me:

TypeError: can only call with other hierarchical index objects

I’ve unstacked the minor index to make it columns, and specified level=’minor’, axis=’columns’ with the same result.

Finally, the end result is to be able to run this same calculation on a DataFrame where the major columns are several equities — in that instance, would .mul() work against each equity as well?

Thanks for your assistance!

Answer 1

Series based it works with level:

df["C01"].mul(s, level=1)

major       minor
2007-09-14  OC       0.745391
2007-09-15  CC       0.480887
            CO       0.325674
            OC       0.753075
            OO       0.505361

Then you can insert it again into your DataFrame. But that should work with DataFrames too, maybe you can suggest it.