Python

Question

我正在尝试使用 pandas_datareader 捕获股票数据。给定少量股票代码，只需手动编写每个 DataReader 调用的脚本，然后连接结果就足够容易了。但是，如果列表的长度增长到更长的数字，则必须有一种更简单的方法来迭代该过程。

import pandas_datareader.data as web
import pandas as pd

symbols = ['AMZN','AAPL','MSFT','NFLX','GOOGL']
for i in symbols:
    print(i)

dfAMZN = web.DataReader('AMZN','yahoo')
dfAMZN['Symbol'] = 'AMZN'
dfAMZN['Date'] = dfAMZN.index

dfAAPL = web.DataReader('AAPL','yahoo')
dfAAPL['Symbol'] = 'AAPL'
dfAAPL['Date'] = dfAAPL.index

dfMSFT = web.DataReader('MSFT','yahoo')
dfMSFT['Symbol'] = 'MSFT'
dfMSFT['Date'] = dfMSFT.index

dfNFLX = web.DataReader('NFLX','yahoo')
dfNFLX['Symbol'] = 'NFLX'
dfNFLX['Date'] = dfNFLX.index

dfGOOGL = web.DataReader('GOOGL','yahoo')
dfGOOGL['Symbol'] = 'GOOGL'
dfGOOGL['Date'] = dfGOOGL.index

frames = [dfAMZN, dfAAPL, dfMSFT, dfNFLX, dfGOOGL]
dfStocks = pd.concat(frames)

有没有一种方法可以遍历符号列表并执行以下步骤，而不仅仅是打印 i？

Answer 1

对新列使用 list comprehension 和 assign：

symbols = ['AMZN','AAPL','MSFT','NFLX','GOOGL']
frames = [web.DataReader(i,'yahoo').assign(Symbol = i, Date = lambda x: x.index) 
          for i in symbols]

dfStocks = pd.concat(frames)

另一种选择：

symbols = ['AMZN','AAPL','MSFT','NFLX','GOOGL']

frames = []
for i in symbols:
    df = web.DataReader(i,'yahoo')
    df['Symbol'] = i
    df['Date'] = df.index
    frames.append(df)

dfStocks = pd.concat(frames)

Python - 使用 DataReader 连接数据帧列表

Python - Concat List of Dataframes using DataReader

dataframe

pandas

pandas-datareader