take a conditional mean in pandas-CodePudding

I have a data set like this:

I want to take the mean of LCR in each year for WFR=1 and WFR=0 separately, for example in 2018 I have 4 WFR=0 so take the mean of LCR for it and for WFR=1 I have just one. any idea? thanks

CodePudding user response：

You can groupby by year and WFR. This will create every combination that exists in your dataset. If all conditions are not in your dataframe you can create a base dataframe with all conditions and merge with it.

df = data_new4.groupby(['CLOSDATE_year', 'WFR'])['LCR'].mean()

to plot the data you can use the seaborn library.

import seaborn as sns

sns.lineplot(data=df, hue='WFR', x='CLOSDATE_year', y='LCR')

CodePudding user response：

Try with groupby:

>>> data.groupby(["CLOSDATE_year", "WFR"])["LCR"].mean()

CLOSDATE_year  WFR
2011           1      268.750000
2012           1      235.533333
2018           0      192.775000
               1      186.000000
Name: LCR, dtype: float64