安装pandas金融数据库
In [3]:
pip install pandas_datareader --upgrade
Defaulting to user installation because normal site-packages is not writeable Looking in indexes: https://mirrors.ustc.edu.cn/pypi/web/simple Collecting pandas_datareaderDownloading https://mirrors.ustc.edu.cn/pypi/packages/3f/16/56c9d648b503619ebe96f726b5f642b68e299b34162ed2d6faa9d7966b7d/pandas_datareader-0.10.0-py3-none-any.whl (109 kB) Requirement already satisfied: lxml in c:\programdata\anaconda3\lib\site-packages (from pandas_datareader) (5.2.1) Requirement already satisfied: pandas>=0.23 in c:\programdata\anaconda3\lib\site-packages (from pandas_datareader) (2.2.2) Requirement already satisfied: requests>=2.19.0 in c:\programdata\anaconda3\lib\site-packages (from pandas_datareader) (2.32.3) Requirement already satisfied: numpy>=1.26.0 in c:\programdata\anaconda3\lib\site-packages (from pandas>=0.23->pandas_datareader) (1.26.4) Requirement already satisfied: python-dateutil>=2.8.2 in c:\programdata\anaconda3\lib\site-packages (from pandas>=0.23->pandas_datareader) (2.9.0.post0) Requirement already satisfied: pytz>=2020.1 in c:\programdata\anaconda3\lib\site-packages (from pandas>=0.23->pandas_datareader) (2024.1) Requirement already satisfied: tzdata>=2022.7 in c:\programdata\anaconda3\lib\site-packages (from pandas>=0.23->pandas_datareader) (2023.3) Requirement already satisfied: charset-normalizer<4,>=2 in c:\programdata\anaconda3\lib\site-packages (from requests>=2.19.0->pandas_datareader) (3.3.2) Requirement already satisfied: idna<4,>=2.5 in c:\programdata\anaconda3\lib\site-packages (from requests>=2.19.0->pandas_datareader) (3.7) Requirement already satisfied: urllib3<3,>=1.21.1 in c:\programdata\anaconda3\lib\site-packages (from requests>=2.19.0->pandas_datareader) (2.2.3) Requirement already satisfied: certifi>=2017.4.17 in c:\programdata\anaconda3\lib\site-packages (from requests>=2.19.0->pandas_datareader) (2024.8.30) Requirement already satisfied: six>=1.5 in c:\programdata\anaconda3\lib\site-packages (from python-dateutil>=2.8.2->pandas>=0.23->pandas_datareader) (1.16.0) Installing collected packages: pandas_datareader Successfully installed pandas_datareader-0.10.0 Note: you may need to restart the kernel to use updated packages.
测试pandas
In [4]:
import pandas_datareader as web
web.__version__
Out[4]:
'0.10.0'
安装tushare国内K线数据源
In [5]:
!pip install tushare
Defaulting to user installation because normal site-packages is not writeable Looking in indexes: https://mirrors.ustc.edu.cn/pypi/web/simple Collecting tushareDownloading https://mirrors.ustc.edu.cn/pypi/packages/33/3e/d426a56e5feac9b0aaada1c6b0745ed03422d4a713295e0bbb44c8ea86fe/tushare-1.4.29-py3-none-any.whl (142 kB) Requirement already satisfied: pandas in c:\programdata\anaconda3\lib\site-packages (from tushare) (2.2.2) Requirement already satisfied: requests in c:\programdata\anaconda3\lib\site-packages (from tushare) (2.32.3) Requirement already satisfied: lxml in c:\programdata\anaconda3\lib\site-packages (from tushare) (5.2.1) Collecting simplejson (from tushare)Downloading https://mirrors.ustc.edu.cn/pypi/packages/b9/c4/90de06b2d8737c68c05ff9274113f854dbf6a5f28b7a955212111672cb57/simplejson-4.1.1-cp312-cp312-win_amd64.whl (90 kB) Collecting bs4 (from tushare)Downloading https://mirrors.ustc.edu.cn/pypi/packages/51/bb/bf7aab772a159614954d84aa832c129624ba6c32faa559dfb200a534e50b/bs4-0.0.2-py2.py3-none-any.whl (1.2 kB) Requirement already satisfied: websocket-client>=0.57.0 in c:\programdata\anaconda3\lib\site-packages (from tushare) (1.8.0) Requirement already satisfied: tqdm in c:\programdata\anaconda3\lib\site-packages (from tushare) (4.66.5) Requirement already satisfied: beautifulsoup4 in c:\programdata\anaconda3\lib\site-packages (from bs4->tushare) (4.12.3) Requirement already satisfied: numpy>=1.26.0 in c:\programdata\anaconda3\lib\site-packages (from pandas->tushare) (1.26.4) Requirement already satisfied: python-dateutil>=2.8.2 in c:\programdata\anaconda3\lib\site-packages (from pandas->tushare) (2.9.0.post0) Requirement already satisfied: pytz>=2020.1 in c:\programdata\anaconda3\lib\site-packages (from pandas->tushare) (2024.1) Requirement already satisfied: tzdata>=2022.7 in c:\programdata\anaconda3\lib\site-packages (from pandas->tushare) (2023.3) Requirement already satisfied: charset-normalizer<4,>=2 in c:\programdata\anaconda3\lib\site-packages (from requests->tushare) (3.3.2) Requirement already satisfied: idna<4,>=2.5 in c:\programdata\anaconda3\lib\site-packages (from requests->tushare) (3.7) Requirement already satisfied: urllib3<3,>=1.21.1 in c:\programdata\anaconda3\lib\site-packages (from requests->tushare) (2.2.3) Requirement already satisfied: certifi>=2017.4.17 in c:\programdata\anaconda3\lib\site-packages (from requests->tushare) (2024.8.30) Requirement already satisfied: colorama in c:\programdata\anaconda3\lib\site-packages (from tqdm->tushare) (0.4.6) Requirement already satisfied: six>=1.5 in c:\programdata\anaconda3\lib\site-packages (from python-dateutil>=2.8.2->pandas->tushare) (1.16.0) Requirement already satisfied: soupsieve>1.2 in c:\programdata\anaconda3\lib\site-packages (from beautifulsoup4->bs4->tushare) (2.5) Installing collected packages: simplejson, bs4, tushare Successfully installed bs4-0.0.2 simplejson-4.1.1 tushare-1.4.29
测试获取贵州茅台的数据
import tushare as ts# ---------------------- 1. 初始化 Tushare ----------------------
# 把这里替换成你自己的 TOKEN
ts.set_token("你的key")
pro = ts.pro_api()# ---------------------- 2. 获取 A 股日线数据 ----------------------
# 示例：贵州茅台 600519.SH
df = pro.daily(ts_code="600519.SH", start_date="20260401", end_date="20260511")# 按交易日期排序
df = df.sort_values("trade_date")# ---------------------- 3. 查看数据 ----------------------
print("===== A股数据前5行 =====")
print(df.head())# 保存数据
#df.to_csv("600519.csv", index=False)
===== A股数据前5行 =====ts_code trade_date open high low close pre_close \ 3 600519.SH 20260506 1365.10 1379.00 1360.05 1375.00 1384.79 2 600519.SH 20260507 1375.00 1388.00 1370.01 1371.05 1375.00 1 600519.SH 20260508 1371.66 1382.77 1370.00 1372.99 1371.05 0 600519.SH 20260511 1372.89 1372.89 1361.00 1361.33 1372.99 change pct_chg vol amount 3 -9.79 -0.7070 47806.04 6550750.940 2 -3.95 -0.2873 40461.47 5573286.315 1 1.94 0.1415 33368.53 4582855.939 0 -11.66 -0.8492 57135.10 7790721.392
测试根据数据画K线图
我们发现tushare接口稳定且返回迅速。可查询时间范围也远超akshare。
import tushare as ts
import matplotlib.pyplot as plt
import pandas as pdts.set_token("你的key")
pro = ts.pro_api()df = pro.daily(ts_code="600519.SH", start_date="20260401", end_date="20260511")
df = df.sort_values("trade_date")# 时间转换（必须加这行，X轴才会正常显示）
df['trade_date'] = pd.to_datetime(df['trade_date'])# 画图：以时间为X轴
plt.figure(figsize=(10,5))
plt.plot(df['trade_date'], df['close'], linewidth=2, color='red')
plt.grid(True)
plt.xticks(rotation=45)
#plt.title("茅台收盘价")
plt.tight_layout()
plt.show()
策略1连续3日机构进入的Top5
由于tushare龙虎榜数据需要购买权限才能访问，此接口调用不成功。可换akshare接口。akshare接口虽然可以调用龙虎榜数据，但日线数据非常不稳定，经常返回'Connection aborted.'
import tushare as ts
ts.set_token("你的key")
pro = ts.pro_api()
df = pro.top_inst(trade_date='20260511')
print(df)
---------------------------------------------------------------------------
Exception Traceback (most recent call last)
Cell In[24], line 4
2 ts.set_token("你的key")
3 pro = ts.pro_api()
----> 4 df = pro.top_inst(trade_date='20260511')
5 print(df)File ~\AppData\Roaming\Python\Python312\site-packages\tushare\pro\client.py:46, in DataApi.query(self, api_name, fields, **kwargs)
44 result = json.loads(res.text)
45 if result['code'] != 0:
---> 46 raise Exception(result['msg'])
47 data = result['data']
48 columns = data['fields']Exception: 抱歉，您没有接口(top_inst)访问权限，权限的具体详情访问：https://tushare.pro/document/1?doc_id=108。
import akshare as ak
import pandas as pd# 获取区间龙虎榜数据
df = ak.stock_lhb_detail_em(start_date="20260506", end_date="20260510")# 确保净买额为数值
df["龙虎榜净买额"] = pd.to_numeric(df["龙虎榜净买额"], errors="coerce")# ===================== 核心逻辑 =====================
# 1. 先按【日期 + 净买额】排序
# 2. 按日期分组，每组只取前 5 名
df_result = (df.sort_values(["上榜日", "龙虎榜净买额"], ascending=[True, False]).groupby("上榜日").head(5) # 每日只保留前5[["上榜日", "代码", "名称", "龙虎榜净买额", "涨跌幅", "上榜原因"]]
)# 打印输出
print("===== 每日龙虎榜净买额 TOP5 =====")
print(df_result.to_string(index=False))
===== 每日龙虎榜净买额 TOP5 =====上榜日 代码 名称 龙虎榜净买额 涨跌幅 上榜原因 2026-05-06 301308 江波龙 1.935484e+09 19.9990 日涨幅达到15%的前5只证券 2026-05-06 000338 潍柴动力 1.131065e+09 10.0000 日涨幅偏离值达到7%的前5只证券 2026-05-06 001309 德明利 5.669867e+08 9.9998 日涨幅偏离值达到7%的前5只证券 2026-05-06 300166 东方国信 4.850638e+08 20.0130 日涨幅达到15%的前5只证券 2026-05-06 601016 节能风电 4.385411e+08 10.0939 有价格涨跌幅限制的日收盘价格涨幅偏离值达到7%的前五只证券 2026-05-07 000066 中国长城 1.155660e+09 10.0000 连续三个交易日内，涨幅偏离值累计达到20%的证券 2026-05-07 000066 中国长城 5.317460e+08 10.0000 日涨幅偏离值达到7%的前5只证券 2026-05-07 002342 巨力索具 4.971131e+08 9.9950 日换手率达到20%的前5只证券 2026-05-07 002342 巨力索具 4.971131e+08 9.9950 日涨幅偏离值达到7%的前5只证券 2026-05-07 301171 易点天下 4.264262e+08 20.0112 日涨幅达到15%的前5只证券 2026-05-08 600522 中天科技 2.040058e+09 6.5659 非ST、*ST和S证券连续三个交易日内收盘价格涨幅偏离值累计达到20%的证券 2026-05-08 600498 烽火通信 1.280115e+09 9.9961 非ST、*ST和S证券连续三个交易日内收盘价格涨幅偏离值累计达到20%的证券 2026-05-08 000547 航天发展 7.214594e+08 10.0080 日涨幅偏离值达到7%的前5只证券 2026-05-08 600589 大位科技 4.965516e+08 10.0264 非ST、*ST和S证券连续三个交易日内收盘价格涨幅偏离值累计达到20%的证券 2026-05-08 603256 宏和科技 4.396742e+08 3.3030 非ST、*ST和S证券连续三个交易日内收盘价格涨幅偏离值累计达到20%的证券
策略2 双移动平均策略
最简单的策略，当5日均线超过10日均线时买入，当5日均线下穿10日均线时卖出。
import tushare as ts
import matplotlib.pyplot as plt
import pandas as pd
import numpy as npdef getDailyData(code, startDate, endDate):# ---------------------- 1. 初始化 Tushare ----------------------# 把这里替换成你自己的 TOKENts.set_token("你的key")pro = ts.pro_api()# ---------------------- 2. 获取 A 股日线数据 ----------------------# 示例：江波龙 301308.SZ# df = pro.daily(ts_code="301308.SZ", start_date="20260401", end_date="20260511")df = pro.daily(ts_code=code, start_date=startDate, end_date=endDate)# 按交易日期排序df = df.sort_values("trade_date")# ---------------------- 3. 查看数据 ----------------------print(df.head())return dfdef drawPlot(df, *args):# 画图：以时间为X轴plt.figure(figsize=(10, 5))# 绘制股价收盘价，并添加图例标签：收盘价plt.plot(df['trade_date'], df['close'], linewidth=2, color='red', label='收盘价')# 循环绘制均线（args 是你传入的均线列名，如 ma5, ma10, svg_1 等）for arg in args:if arg in df.columns:# 绘制均线，并使用列名作为图例标签plt.plot(df['trade_date'], df[arg], '--', linewidth=2, label=arg)# ========== 添加图表标注信息 ==========plt.title('股价走势与均线图', fontsize=14, fontproperties='SimHei') # 图表标题plt.xlabel('交易日期', fontsize=12, fontproperties='SimHei') # X轴标签plt.ylabel('价格', fontsize=12, fontproperties='SimHei') # Y轴标签# 显示图例（必须有 label 才会显示）plt.legend(prop={'family': 'SimHei'}) # 支持中文图例# 网格 + 日期旋转plt.grid(True)plt.xticks(rotation=45)plt.tight_layout()plt.show()def strategy1(df):#双移动平均策略#复制一份用作策略计算strategy=pd.DataFrame(index=df.index)strategy['date']=df['trade_date']#添加一个signal字段，用来存储交易信号strategy['signal']=0#将5日均线存到avg_5strategy['avg_5']=df['close'].rolling(5).mean()#将10日均线存到avg_10strategy['avg_10']=df['close'].rolling(10).mean()#计算买入信号，当5日均值大于10日均值时设为1 反之为0strategy['signal']=np.where(strategy['avg_5']>strategy['avg_10'],1,0)#根据交易信号变化下单，当交易信号从0变为1时买入，当从1变为0时卖出strategy['order']=strategy['signal'].diff()print("--------------双移动平均策略-----------")print(strategy)return strategydef strategytest(cash, df):# 回测# 执行双移动平均策略strategy = strategy1(df)# 初始启动资金initial_cash = cash# 新建一个 strategy 的拷贝 仓位表positions = pd.DataFrame(index=strategy.index)positions['stock'] = 0 # 初始持仓为 0# 设置最小买入 100 股positions['stock'] = strategy['signal'] * 100# 创建投资组合数据库portfolio = pd.DataFrame(index=strategy.index)# 持仓市值 = 持股数量 × 收盘价portfolio['stock value'] = positions['stock'] * df['close']# 仓位变化 = 下单的数量order = positions.diff()# 剩余资金 = 初始资金 - 累计下单金额portfolio['cash'] = initial_cash - (order['stock'] * df['close']).cumsum()# 总资产 = 现金 + 持仓市值portfolio['total'] = portfolio['cash'] + portfolio['stock value']# 打印结果print("--------------回测结果-----------")print(portfolio)return portfolio # 可选：返回结果方便后续画图if __name__ == '__main__':# 获取股票数据df = getDailyData('600276.SH', '20260401', '20260501')# 增加10日均线列df['avg_10'] = df['close'].rolling(10).mean()df['avg_5'] = df['close'].rolling(5).mean()# 画出来drawPlot(df, 'avg_10', 'avg_5')#执行双移动平均策略回测strategytest(20000,df)
ts_code trade_date open high low close pre_close change \ 20 600276.SH 20260401 56.03 57.78 55.60 57.73 55.22 2.51 19 600276.SH 20260402 57.74 58.35 57.01 57.37 57.73 -0.36 18 600276.SH 20260403 56.98 56.99 56.10 56.38 57.37 -0.99 17 600276.SH 20260407 56.51 56.85 55.00 56.12 56.38 -0.26 16 600276.SH 20260408 57.01 57.64 55.91 57.45 56.12 1.33 pct_chg vol amount 20 4.5455 1005516.56 5750559.322 19 -0.6236 765169.85 4415860.425 18 -1.7256 454916.05 2569466.951 17 -0.4612 401466.88 2244163.329 16 2.3699 734462.90 4189267.319
--------------双移动平均策略-----------
date signal avg_5 avg_10 order 20 20260401 0 NaN NaN NaN 19 20260402 0 NaN NaN 0.0 18 20260403 0 NaN NaN 0.0 17 20260407 0 NaN NaN 0.0 16 20260408 0 57.010 NaN 0.0 15 20260409 0 56.856 NaN 0.0 14 20260410 0 56.794 NaN 0.0 13 20260413 0 56.558 NaN 0.0 12 20260414 0 56.570 NaN 0.0 11 20260415 0 56.626 56.818 0.0 10 20260416 0 56.728 56.792 0.0 9 20260417 0 56.582 56.688 0.0 8 20260420 1 56.810 56.684 1.0 7 20260421 1 56.724 56.647 0.0 6 20260422 0 56.266 56.446 -1.0 5 20260423 0 56.052 56.390 0.0 4 20260424 0 55.996 56.289 0.0 3 20260427 0 55.804 56.307 0.0 2 20260428 0 55.622 56.173 0.0 1 20260429 0 55.532 55.899 0.0 0 20260430 0 55.028 55.540 0.0 --------------回测结果-----------stock value cash total 20 0.0 NaN NaN 19 0.0 20000.0 20000.0 18 0.0 20000.0 20000.0 17 0.0 20000.0 20000.0 16 0.0 20000.0 20000.0 15 0.0 20000.0 20000.0 14 0.0 20000.0 20000.0 13 0.0 20000.0 20000.0 12 0.0 20000.0 20000.0 11 0.0 20000.0 20000.0 10 0.0 20000.0 20000.0 9 0.0 20000.0 20000.0 8 5634.0 14366.0 20000.0 7 5575.0 14366.0 19941.0 6 0.0 19910.0 19910.0 5 0.0 19910.0 19910.0 4 0.0 19910.0 19910.0 3 0.0 19910.0 19910.0 2 0.0 19910.0 19910.0 1 0.0 19910.0 19910.0 0 0.0 19910.0 19910.0
从回测结果可知，投入20000元，买入点在第8条，卖出点在第7条，结果是亏损90元。
本文介绍了使用python进行股票策略设计、回测验证的过程，只是技术学习交流，不可用于实战。