<p>如果我正确理解你想要达到的目标,你可以尝试以下方法:</p>
<pre><code>import pandas as pd
data = {'household_key':[1,1,1,1,2,2,2,3,3,3],
'age_group':[25,25,25,25,30,30,30,25,25,25],
'income_group':[40,40,40,40,40,40,40,30,30,30],
'day':['2019-01-01','2019-01-05','2019-01-08','2019-01-15','2019-01-01','2019-01-08','2019-01-10','2019-01-01','2019-01-05','2019-01-10']}
df = pd.DataFrame(data)
# get group by household
group1 = df.groupby(['household_key', 'age_group']).agg({'day': 'nunique'})
# get group by age_group
group2 = df.groupby(['age_group']).agg({'day': 'nunique'})
# join the results
group = group2.merge(group1, how='right', left_index=True, right_index=True)
group.columns = ['unique_days_in_group', 'unique_days_in_household']
print(group)
</code></pre>
<p>结果如下:</p>
<pre><code> unique_days_in_group unique_days_in_household
household_key age_group
1 25 5 4
2 30 3 3
3 25 5 3
</code></pre>