Python 结合 Merge 填充 Pandas Dataframe 中的缺失值

有一种方法可以做到这一点(取决于问题中的列命名方案，如果真正的数据帧不使用这些名称，则需要进行调整):

# get a list of var names
vars = [col for col in df.columns if col.startswith('var')]
# make a dataframe with all the unique dates as its index
dates = pd.DataFrame(pd.concat([df[f'date_{v}'] for v in vars]).unique()).set_index(0)
# extract each of the date_var* and var* columns to a separate df
# and join to the list of dates, filling na with 0
dfs = [dates.join(df[[f'date_{v}', v]].set_index(f'date_{v}')).fillna(0) for v in vars]
# concatenate all the dfs together and reset the index
out = pd.concat(dfs, axis=1).reset_index(names='date')

输出(针对您的示例数据):

       date  var1  var2
0  1/1/2022  11.0  12.0
1  1/2/2022  21.0  22.0
2  1/3/2022  31.0   0.0
3  1/4/2022  41.0  42.0
4  1/5/2022   0.0  51.0

Edit个

根据对问题的更改，对代码的此修改应执行所需的操作:

ticker_cols = [col for col in df.columns if col.endswith('Return')]
date_cols = [col for col in df.columns if col.startswith('Date')]
dates = pd.DataFrame(pd.concat([df[date] for date in date_cols]).unique()).set_index(0)
dfs = [dates.join(df[[date, ticker]].set_index(date)).fillna(0) for date, ticker in zip(date_cols, ticker_cols)]
out = pd.concat(dfs, axis=1).reset_index(names='date')

输出:

       date  ABC Return  XYZ Return
0  1/1/2022        11.0        12.0
1  1/2/2022        21.0        22.0
2  1/3/2022        31.0         0.0
3  1/4/2022        41.0        42.0
4  1/5/2022         0.0        51.0