如何修复Pandas的SettingWithCopyWarning警告（非忽略）

问题背景 #

我通过yfinance库从Yahoo Finance获取股票交易的价格与成交量数据，创建了包含收盘价（Close）和成交量（Volume）的Closes DataFrame。尝试为该DataFrame添加年、季、月、周维度的成交量均值列时，触发了SettingWithCopyWarning警告。希望彻底消除该警告，同时保证新增列的计算逻辑正确，而非简单忽略警告。

原代码 #

import yfinance as yf
import pandas as pd
import warnings
warnings.filterwarnings("ignore", message="The 'unit' keyword in TimedeltaIndex construction is deprecated and will be removed in a future version. Use pd.to_timedelta instead.", category=FutureWarning, module="yfinance.utils")
warnings.filterwarnings("ignore", category=pd.errors.PerformanceWarning)
#warnings.filterwarnings("ignore", message="DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`", category=pd.errors.PerformanceWarning)

TickersList=['A', 'AAL', 'AAPL', 'ABBV', 'ABNB', 'ABT', 'ACGL', 'ACN', 'ADBE', 'ADI', 'ADM']
Stocks=yf.download(TickersList[0::1], period="1y", interval="1d", group_by='ticker')
Stocks.sort_index(level=0,axis=1,inplace=True)
Closes=Stocks.loc[:, (slice(None), ['Close', 'Volume'])]
for i in Closes.columns.get_level_values(0): 
    Closes.loc[:,(i,'Meam1Y')]=Closes.loc[:,(i,'Volume')].rolling (250).mean() 
    Closes.loc[:,(i,'Meam1Q')]=Closes.loc[:,(i,'Volume')].rolling (62).mean() 
    Closes.loc[:,(i,'Meam1M')]=Closes.loc[:,(i,'Volume')].rolling (20).mean() 
    Closes.loc[:,(i,'Meam1W')]=Closes.loc[:,(i,'Volume')].rolling (5).mean() 
Closes

警告内容 #

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  Closes.loc[:,(i,'Meam1Y')]=Closes.loc[:,(i,'Volume')].rolling (250).mean() 
C:\Users\iiiva\AppData\Local\Temp\ipykernel_27188\26776804.py:4: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

解决方案 #

核心原因 #

Closes是从Stocks切片得到的视图而非独立副本，直接修改视图会触发Pandas的SettingWithCopyWarning——因为Pandas无法确定你是想修改原数据还是副本。

修复步骤 #

1. 显式创建独立副本
   在生成Closes时，调用.copy()方法，让Closes成为完全独立的DataFrame，脱离与原Stocks的关联：

   Closes = Stocks.loc[:, (slice(None), ['Close', 'Volume'])].copy()
   

2. 可选：优化列添加逻辑（提升性能）
   循环逐个添加列会导致DataFrame碎片化，虽然你已忽略PerformanceWarning，但可以通过批量生成新列后合并的方式优化性能：

   # 预先存储所有新列
   new_columns = []
   for i in Closes.columns.get_level_values(0):
       col_y = Closes.loc[:, (i, 'Volume')].rolling(250).mean().rename((i, 'Meam1Y'))
       col_q = Closes.loc[:, (i, 'Volume')].rolling(62).mean().rename((i, 'Meam1Q'))
       col_m = Closes.loc[:, (i, 'Volume')].rolling(20).mean().rename((i, 'Meam1M'))
       col_w = Closes.loc[:, (i, 'Volume')].rolling(5).mean().rename((i, 'Meam1W'))
       new_columns.extend([col_y, col_q, col_m, col_w])
   # 合并所有新列到Closes
   Closes = pd.concat([Closes] + new_columns, axis=1)
   

修改后完整代码 #

import yfinance as yf
import pandas as pd
import warnings
warnings.filterwarnings("ignore", message="The 'unit' keyword in TimedeltaIndex construction is deprecated and will be removed in a future version. Use pd.to_timedelta instead.", category=FutureWarning, module="yfinance.utils")
warnings.filterwarnings("ignore", category=pd.errors.PerformanceWarning)

TickersList=['A', 'AAL', 'AAPL', 'ABBV', 'ABNB', 'ABT', 'ACGL', 'ACN', 'ADBE', 'ADI', 'ADM']
Stocks=yf.download(TickersList[0::1], period="1y", interval="1d", group_by='ticker')
Stocks.sort_index(level=0,axis=1,inplace=True)
# 显式创建副本
Closes=Stocks.loc[:, (slice(None), ['Close', 'Volume'])].copy()

# 批量生成并合并新列
new_columns = []
for i in Closes.columns.get_level_values(0):
    col_y = Closes.loc[:, (i, 'Volume')].rolling(250).mean().rename((i, 'Meam1Y'))
    col_q = Closes.loc[:, (i, 'Volume')].rolling(62).mean().rename((i, 'Meam1Q'))
    col_m = Closes.loc[:, (i, 'Volume')].rolling(20).mean().rename((i, 'Meam1M'))
    col_w = Closes.loc[:, (i, 'Volume')].rolling(5).mean().rename((i, 'Meam1W'))
    new_columns.extend([col_y, col_q, col_m, col_w])

Closes = pd.concat([Closes] + new_columns, axis=1)
Closes

效果验证 #

修改后运行代码，SettingWithCopyWarning会彻底消失，同时新增列的计算逻辑与原代码完全一致，还能避免DataFrame碎片化带来的性能问题。

内容的提问来源于stack exchange，提问作者Vasiliy Deryuga