python金融数据分析和可视化--03利用Akshare获取股票数据
02利用Akshare获取股票数据
1. AKShare 的介绍
AKShare 是基于 Python 的财经数据接口库，目的是实现对股票、期货、期权、基金、外汇、债券、指数、加密货币等金融产品的基本面数据、实时和历史行情数据、衍生数据从数据采集、数据清洗到数据落地的一套工具，主要用于学术研究目的。
AKShare 的特点是获取的是相对权威的财经数据网站公布的原始数据，通过利用原始数据进行各数据源之间的交叉验证，进而再加工，从而得出科学的结论。
2. 安装 AKShare
pip install akshare
3. 获取股票数据
AKShare 股票数据
AKShare github
# 股票市场总貌
import akshare as ak
# 上海证券交易所
# http://www.sse.com.cn/market/stockdata/statistic/
def sh_df():
stock_sse_summary_df = ak.stock_sse_summary()
print(stock_sse_summary_df)
# 深圳证券交易所
# 证券类别统计
# http://www.szse.cn/market/overview/index.html
def sz_df():
stock_szse_summary_df = ak.stock_szse_summary()
print(stock_szse_summary_df)
# 深圳证券交易所
# 地区交易排序
# http://www.szse.cn/market/overview/index.html
def sz_area():
stock_szse_area_summary_df = ak.stock_szse_area_summary(date="202203")
print(stock_szse_area_summary_df)
# 深圳证券交易所
# 股票行业成交
# http://docs.static.szse.cn/www/market/periodical/month/W020220511355248518608.html
def sz_sector():
stock_szse_sector_summary_df = ak.stock_szse_sector_summary(symbol="当年", date="202204")
print(stock_szse_sector_summary_df)
# 上海证券交易所
# 每日概况
# http://www.sse.com.cn/market/stockdata/overview/day/
def sh_day():
stock_sse_deal_daily_df = ak.stock_sse_deal_daily(date="20201111")
print(stock_sse_deal_daily_df)
def get_account_statistics():
# 股票账户统计月度
"""
输出参数
名称 类型 描述
数据日期 object -
新增投资者-数量 float64 注意单位: 万户
新增投资者-环比 float64 -
新增投资者-同比 float64 -
期末投资者-总量 float64 注意单位: 万户
期末投资者-A股账户 float64 注意单位: 万户
期末投资者-B股账户 float64 注意单位: 万户
沪深总市值 float64 -
沪深户均市值 float64 注意单位: 万
上证指数-收盘 float64 -
上证指数-涨跌幅 float64 -
"""
account = ak.stock_account_statistics_em()
account.set_index("数据日期", inplace=True) # 设置索引值
account.to_csv("I:\\bianchengxx\\pythonxx\\backtrader_001\\datas\\stock_account_statistics.csv")
print(account)
if __name__ == '__main__':
# sh_df()
# sz_df()
# sz_area()
# sz_sector()
# sh_day()
get_account_statistics()
# 个股信息查询
# http://quote.eastmoney.com/concept/sh603777.html?from=classic
import akshare as ak
df = ak.stock_individual_info_em(symbol="002624")
print(df)
# 实时行情数据-东财
# 沪深京 A 股
# http://quote.eastmoney.com/center/gridlist.html#hs_a_board
import akshare as ak
# 实时行情数据-东财
# 沪深京 A 股
# 单次返回所有沪深京 A 股上市公司的实时行情数据
def em_spot():
stock_zh_a_spot_em_df = ak.stock_zh_a_spot_em()
print(stock_zh_a_spot_em_df)
# 实时行情数据-东财
# 沪 A 股
# http://quote.eastmoney.com/center/gridlist.html#sh_a_board
def em_sha_spot():
stock_sh_a_spot_em_df = ak.stock_sh_a_spot_em()
print(stock_sh_a_spot_em_df)
# 实时行情数据-东财
# 深 A 股
# http://quote.eastmoney.com/center/gridlist.html#sz_a_board
def em_sza_spot():
stock_sz_a_spot_em_df = ak.stock_sz_a_spot_em()
print(stock_sz_a_spot_em_df)
# 实时行情数据-东财
# 京 A 股
# http://quote.eastmoney.com/center/gridlist.html#bj_a_board
def em_bja_spot():
stock_bj_a_spot_em_df = ak.stock_bj_a_spot_em()
print(stock_bj_a_spot_em_df)
# 实时行情数据-东财
# 新股
# http://quote.eastmoney.com/center/gridlist.html#newshares
def em_new_spot():
stock_new_a_spot_em_df = ak.stock_new_a_spot_em()
print(stock_new_a_spot_em_df)
# 实时行情数据-东财
# 科创板
# http://quote.eastmoney.com/center/gridlist.html#hs_a_board
def em_kc_spot():
stock_kc_a_spot_em_em_df = ak.stock_kc_a_spot_em()
print(stock_kc_a_spot_em_em_df)
# 实时行情数据-新浪
# http://vip.stock.finance.sina.com.cn/mkt/#hs_a
def xl_a_spot():
stock_zh_a_spot_df = ak.stock_zh_a_spot()
print(stock_zh_a_spot_df)
if __name__ == "__main__":
em_spot()
# em_sha_spot()
# em_sza_spot()
# 历史行情数据-东财
# https://quote.eastmoney.com/concept/sh603777.html?from=classic
import akshare as ak
import pandas as pd
pd.set_option('expand_frame_repr', False) # True就是可以换行显示。设置成False的时候不允许换行
pd.set_option('display.max_columns', None) # 显示所有列
# pd.set_option('display.max_rows', None) # 显示所有行
pd.set_option('colheader_justify', 'centre') # 显示居中
def a_hist():
# period
# str
# choice of {'daily', 'weekly', 'monthly'}
# start_date
# str
# 开始查询的日期
# end_date
# str
# 结束查询的日期
# adjust
# str
# 默认返回不复权的数据;
# qfq: 返回前复权后的数据;
# hfq: 返回后复权后的数据
stock_zh_a_hist_df = ak.stock_zh_a_hist(symbol="002624", period="weekly", start_date="20190301", end_date='20230907',
adjust="qfq")
print(stock_zh_a_hist_df)
# 分时数据-东财
# http://quote.eastmoney.com/concept/sh603777.html?from=classic
# period str choice of {'1', '5', '15', '30', '60'};
# adjust str choice of {'', 'qfq', 'hfq'}; '': 不复权, 'qfq': 前复权, 'hfq': 后复权,
# 其中 1 分钟数据返回近 5 个交易日数据且不复权
def a_hist_min():
stock_zh_a_hist_min_em_df = ak.stock_zh_a_hist_min_em(symbol="002624", start_date="2023-01-01 09:30:00", end_date="2023-02-03 15:00:00", period='30', adjust='')
print(stock_zh_a_hist_min_em_df)
if __name__ == "__main__":
a_hist()
# a_hist_min()
4. 获取股票数据本地存储
import akshare as ak
import pandas as pd
pd.set_option('expand_frame_repr', False) # True就是可以换行显示。设置成False的时候不允许换行
pd.set_option('display.max_columns', None) # 显示所有列
pd.set_option('display.max_rows', None) # 显示所有行
pd.set_option('colheader_justify', 'centre') # 显示居中
def download_hist(symbol="002624", period="daily", start="1990101", end="20230318", adjust="qfq"):
stock_zh_a_hist_df = ak.stock_zh_a_hist(symbol=symbol, period=period, start_date=start, end_date=end,
adjust=adjust)
stock_zh_a_hist_df.sort_values("日期", inplace=True)
stock_zh_a_hist_df.set_index("日期", inplace=True)
if period=="daily":
stock_zh_a_hist_df.to_csv("I:\\akshare_stock\\stock_datas\\day\\{}.csv".format(symbol))
elif period=="weekly":
stock_zh_a_hist_df.to_csv("I:\\akshare_stock\\stock_datas\\week\\{}.csv".format(symbol))
elif period=="monthly":
stock_zh_a_hist_df.to_csv("I:\\akshare_stock\\stock_datas\\month\\{}.csv".format(symbol))
else:
print("period错误")
def download_hist_min(symbol="002624", start="1990-01-01 09:30:00", end="2023-02-03 15:00:00", period='30', adjust='qfq'):
stock_zh_a_hist_min_em_df = ak.stock_zh_a_hist_min_em(symbol=symbol, start_date=start, end_date=end, period=period, adjust=adjust)
stock_zh_a_hist_min_em_df.set_index("时间", inplace=True)
stock_zh_a_hist_min_em_df.to_csv("I:\\akshare_stock\\stock_datas\\minute\\"+symbol+"_{}.csv".format(period))
if __name__ == "__main__":
df = pd.read_csv("I:\\akshare_stock\\stock_datas\\stock_list.csv")
df.sort_values("symbol", inplace=True)
code = list(df["ts_code"])
print(len(code))
for period in ['daily', 'weekly', 'monthly']:
for i in range(0, len(code)):
symbol = code[i].rstrip('.SZHBJ')
download_hist(symbol=symbol, period=period)
print(symbol)
for period in ['1', '5', '15', '30', '60']:
for i in range(0, len(code)):
symbol = code[i].rstrip('.SZHBJ')
download_hist_min(symbol=symbol, period=period)
print(symbol)
5.将股票数据存到mysql数据库中
"""
date:20210918
将CSV文件写入到MySQL中
"""
import pandas as pd
from sqlalchemy import create_engine
def connect_db(db):
engine = create_engine('mysql+pymysql://hao:671010@localhost:3306/{}?charset=utf8'.format(db))
return engine
def create_stock(akcode, db, date):
# 读取本地CSV文件
df = pd.read_csv(
'I:\\akshare_stock\\stock_datas\\' + date + '\\{}.csv'.format(akcode))
engine = connect_db(db)
# name='stocklist'全部小写否则会报错
df.to_sql(name='ak_' + date + '_{}'.format(akcode), con=engine, index=False, if_exists='replace')
def create_min_stock(akcode, db, date):
# 读取本地CSV文件
df = pd.read_csv(
'I:\\akshare_stock\\stock_datas\\minute\\'+akcode+'_{}.csv'.format(date))
engine = connect_db(db)
# name='stocklist'全部小写否则会报错
df.to_sql(name='ak_' + akcode + '_{}'.format(date), con=engine, index=False, if_exists='replace')
def read_csv(code):
df = pd.read_csv('I:\\akshare_stock\\stock_datas\\{}.csv'.format(code))
return df
stockDB = 'akshare_stock'
stockList = 'stock_list'
create_stock(akcode=stockList, db=stockDB, date="day")
df1 = read_csv(stockList)
df1.sort_values("ts_code", inplace=True)
li1 = list(df1['ts_code'])
for date in ["day", "week", "month"]:
for i in range(0, len(li1)):
codeStock = li1[i].rstrip('.SHZBJ')
create_stock(akcode=codeStock, db=stockDB, date=date)
print(codeStock)
for date in ['1', '5', '15', '30', '60']:
for i in range(0, len(li1)):
codeStock = li1[i].rstrip('.SHZBJ')
create_min_stock(akcode=codeStock, db=stockDB, date=date)
print(codeStock)