Pandas Groupby Aggregation Multiple Compute Functions

Often you may need to aggregate data such that you compute multiple functions against all columns of data. In this example, I demonstrate how to aggregate data with pandas groupby using multiple compute methods.

import numpy as np
import pandas as pd
from datetime import datetime

To begin, let's read a file with some financial data at the daily level for multiple stocks.

sample_df = pd.read_csv('full_data.csv')
sample_df.head()
date Ticker 1. open 2. high 3. low 4. close 5. volume
0 2019-07-03 AAPL 203.603278 203.653537 203.560323 203.606995 53458.197115
1 2019-07-03 AXP 125.587369 125.616468 125.558919 125.587030 14894.287805
2 2019-07-03 BA 353.382682 353.453835 353.301135 353.377761 32912.900474
3 2019-07-03 CAT 135.110770 135.152305 135.065153 135.105175 10665.087379
4 2019-07-03 CSCO 56.063977 56.083697 56.045637 56.065251 61533.024038

Below, we perform multiple aggregations on all price and volume columns to return the mean and var by day.

sample_df.groupby(['date', 'Ticker'], as_index=False).agg([np.mean, np.var]).head()
1. open 2. high 3. low 4. close 5. volume
mean var mean var mean var mean var mean var
date Ticker
2019-07-03 AAPL 203.603278 0.095705 203.653537 0.091135 203.560323 0.104315 203.606995 0.099121 53458.197115 1.364074e+10
AXP 125.587369 0.073431 125.616468 0.072509 125.558919 0.075973 125.587030 0.074976 14894.287805 1.688696e+08
BA 353.382682 0.272969 353.453835 0.273478 353.301135 0.272263 353.377761 0.272168 32912.900474 2.697468e+10
CAT 135.110770 0.169655 135.152305 0.173295 135.065153 0.155551 135.105175 0.161341 10665.087379 7.142522e+07
CSCO 56.063977 0.091510 56.083697 0.089896 56.045637 0.094455 56.065251 0.092519 61533.024038 2.410866e+10