dataframe多重行索引怎么再插入一行

Question

如下图：我想在每个location下面的date里再插入一行，计算6月2日与1日的差值

Answer 1

def insert(df):
    level_0 = df.index.values[0][0] # 取得第一列索引名
    diff = df.loc[(level_0, '2020-06-02'), 'new_cases'] - df.loc[(level_0, '2020-06-01'), 'new_cases'] # 计算差值
    df.loc[(level_0, 'diff'), 'new_cases'] = diff # 插入值
    return df

data.groupby('location').apply(insert).droplevel(0)