如何使用 yfinance 下载的数据将多个股票代码分离到单独的数据帧中

Question

我正在尝试使用 yfinance 下载股票数据信息。目前，我可以使用

yf.download

成功下载单个股票代码，它返回一个包含我可以使用的信息的数据帧。此 API 还支持通过单个 API 调用一次下载多个股票代码。由于下载 API 的速率限制，我需要一次查询所有股票代码，而不是调用每个股票代码。

返回的数据帧将所有信息聚集在一起。如何将

multiple_df

数据帧分成单独的数据帧，就像拉动单个股票一样？我已经研究过如何处理使用 yfinance 下载的多级列名称，但它并没有像使用单个代码调用 API 那样构造单独的数据帧。

GOOGL 的单一股票代码示例

            Adj Close   Close    High     Low    Open    Volume
Date
2022-12-13      95.28   95.63   99.53   95.03   97.76  40593700
2022-12-14      94.73   95.07   96.87   93.60   95.20  28733600
2022-12-15      90.53   90.86   93.64   90.01   93.13  40107000
2022-12-16      89.93   90.26   91.33   89.52   90.76  58011800
2022-12-19      88.12   88.44   90.56   88.21   90.26  29493000
...               ...     ...     ...     ...     ...       ...
2024-12-06     174.51  174.71  175.08  171.86  172.03  21462400
2024-12-09     175.37  175.37  176.26  173.65  173.96  25389600
2024-12-10     185.17  185.17  186.36  181.05  182.85  54813000
2024-12-11     195.40  195.40  195.61  184.85  185.31  67894100
2024-12-12     191.96  191.96  195.18  191.71  194.96  32460315

多个下载数据帧的当前输出

Ticker            NVDA                                                  ...        MSFT
Price             Open        High         Low       Close   Adj Close  ...        High         Low       Close   Adj Close    Volume
Date                                                                    ...
2022-12-13   18.531000   18.790001   17.711000   18.072001   18.058865  ...  263.920013  253.070007  256.920013  252.771729  42196900
2022-12-14   17.987000   18.252001   17.420000   17.674000   17.661152  ...  262.589996  254.309998  257.220001  253.066925  35410900
2022-12-15   17.146000   17.320000   16.700001   16.952000   16.939678  ...  254.199997  247.339996  249.009995  244.989441  35560400
2022-12-16   16.864000   17.041000   16.410000   16.570999   16.558958  ...  249.839996  243.509995  244.690002  240.739197  86102000
2022-12-19   16.572001   16.608999   16.145000   16.254000   16.242184  ...  245.210007  238.710007  240.449997  236.567657  29696400
...                ...         ...         ...         ...         ...  ...         ...         ...         ...         ...       ...
2024-12-06  144.600006  145.699997  141.309998  142.440002  142.440002  ...  446.100006  441.769989  443.570007  443.570007  18821000
2024-12-09  138.970001  139.949997  137.130005  138.809998  138.809998  ...  448.329987  440.500000  446.019989  446.019989  19144400
2024-12-10  139.009995  141.820007  133.789993  135.070007  135.070007  ...  449.619995  441.600006  443.329987  443.329987  18469500
2024-12-11  137.360001  140.169998  135.210007  139.309998  139.309998  ...  450.350006  444.049988  448.989990  448.989990  19200200
2024-12-12  137.100006  138.410004  135.800003  137.339996  137.339996  ...  456.164795  449.130005  449.559998  449.559998  20719891

每个股票行情数据帧分开的所需输出

separated_df = {'AAPL': aapl_df,
                'MSFT': msft_df,
                'NVDA': nvda_df}

MultiIndex 数据框还有额外的

names=['Price', 'Ticker']

列，我似乎无法摆脱

代码

import yfinance as yf
import pandas as pd

single = ['GOOGL']

df = yf.download(single, 
                 period='2y', 
                 interval='1d',
                 multi_level_index=False)

print(df)

tickers = ['AAPL', 'MSFT', 'NVDA']
multiple_df = yf.download(tickers, 
                          period='2y', 
                          interval='1d',
                          group_by='Ticker',
                          multi_level_index=False)

print(multiple_df)

Answer 1

好的知道怎么做了。 在 MultiIndex 上建立索引

multiple_df.index.name = None
data = {idx: gp.T for idx, gp in multiple_df.T.groupby(level=0)}

separated_df = {}
for ticker in data:
    new_df = data[ticker][ticker]['Close High Low Open Volume'.split()]
    new_df.columns.name = None
    adj_close = data[ticker][ticker]['Adj Close']
    new_df.insert(0, 'Adj Close', adj_close)
    
    separated_df[ticker] = new_df

print(separated_df)