跳到主要内容

量化工具与平台

本节介绍量化交易中常用的开源工具、商业平台和数据源。

开源框架​

Python生态​

Backtrader​

import backtrader as bt

class MyStrategy(bt.Strategy):
    params = (('period', 20),)

    def __init__(self):
        self.sma = bt.indicators.SimpleMovingAverage(period=self.p.period)

    def next(self):
        if self.data.close > self.sma:
            self.buy()
        elif self.data.close < self.sma:
            self.sell()

# 运行回测
cerebro = bt.Cerebro()
cerebro.addstrategy(MyStrategy)
data = bt.feeds.YahooFinanceData(dataname='AAPL', fromdate=datetime(2020, 1, 1))
cerebro.adddata(data)
cerebro.run()
cerebro.plot()

特点: 灵活、文档完善、社区活跃

VectorBT​

import vectorbt as vbt

# 超快速回测
price = vbt.YFData.download('BTC-USD').get('Close')
fast_ma = vbt.MA.run(price, 10)
slow_ma = vbt.MA.run(price, 50)

entries = fast_ma.ma_crossed_above(slow_ma)
exits = fast_ma.ma_crossed_below(slow_ma)

portfolio = vbt.Portfolio.from_signals(price, entries, exits, freq='1D')
print(portfolio.stats())

特点: 向量化、速度极快、适合大规模参数扫描

Qlib (微软)​

import qlib
from qlib.data import D
from qlib.contrib.model.gbdt import LGBModel

# 初始化
qlib.init(provider_uri='~/.qlib/qlib_data/cn_data')

# 获取数据
instruments = D.instruments(market='csi300')
fields = ['$close', '$volume', 'Ref($close, 1)', 'Mean($close, 20)']
df = D.features(instruments, fields, start_time='2020-01-01', freq='day')

# 训练模型
model = LGBModel()
model.fit(dataset)

特点: 端到端、AI导向、因子挖掘工具

国内框架​

vn.py​

from vnpy.event import EventEngine
from vnpy.trader.engine import MainEngine
from vnpy.trader.ui import MainWindow, create_qapp
from vnpy_ctp import CtpGateway

# 启动事件引擎
event_engine = EventEngine()
main_engine = MainEngine(event_engine)
main_engine.add_gateway(CtpGateway)

# 连接CTP
main_engine.connect(
    setting={
        "用户名": "",
        "密码": "",
        "经纪商代码": "",
        "交易服务器": "",
        "行情服务器": ""
    },
    gateway_name="CTP"
)

特点: 全功能、支持国内全市场、实盘 ready

商业平台​

国内平台​

平台特色费用
聚宽 (JoinQuant)数据丰富、社区活跃免费+付费
米筐 (RiceQuant)专业级、API完善免费+付费
优矿 (UQER)通联数据、研究工具免费+付费
BigQuantAI导向、可视化免费+付费
果仁网非编程、因子平台免费+付费

国际平台​

平台特色费用
QuantConnect多语言、云端、实盘免费+付费
Quantopian已停止运营-
TradingView图表强大、Pine脚本免费+付费
Alpaca零佣金、API友好免费

数据源​

A股数据​

Tushare​

import tushare as ts

pro = ts.pro_api('your_token')

# 日线数据
df = pro.daily(ts_code='000001.SZ', start_date='20230101', end_date='20231231')

# 财务数据
df = pro.income(ts_code='000001.SZ', period='20231231')

# 实时行情
df = ts.pro_bar(ts_code='000001.SZ', pro_api=pro, freq='1min')

覆盖: 股票、基金、期货、宏观、新闻 费用: 积分制，高级数据需付费

AKShare​

import akshare as ak

# 股票历史数据
df = ak.stock_zh_a_hist(symbol="000001", period="daily", start_date="20230101")

# 实时行情
df = ak.stock_zh_a_spot_em()

# 期货数据
df = ak.futures_zh_realtime(symbol="沪深300股指期货主力")

# 期权数据
df = ak.option_cffex_sz50_list()

特点: 完全免费、更新及时、纯Python

其他数据源​

来源覆盖特点
Yahoo Finance全球股票免费、Python库(yfinance)
Alpha Vantage股票、外汇、加密API限制、免费层级
Quandl金融、经济、另类部分免费、机构级
Wind/同花顺A股专业数据机构付费
聚宽/米筐数据A股、因子平台内免费

开发工具​

环境与IDE​

# 推荐Python环境管理
conda create -n quant python=3.11
conda activate quant

# 核心包
pip install pandas numpy numba backtrader vectorbt
pip install akshare tushare
pip install matplotlib plotly
工具用途
VS Code通用IDE，插件丰富
PyCharmPython专业IDE
Jupyter Lab交互式研究
Docker环境隔离与部署

数据库​

TimescaleDB (时序)​

-- 创建 hypertable
CREATE TABLE stock_prices (
    time TIMESTAMPTZ NOT NULL,
    symbol TEXT NOT NULL,
    open DOUBLE PRECISION,
    high DOUBLE PRECISION,
    low DOUBLE PRECISION,
    close DOUBLE PRECISION,
    volume DOUBLE PRECISION
);

SELECT create_hypertable('stock_prices', 'time');

-- 查询
SELECT time_bucket('1 day', time) AS day,
       first(open, time) as open,
       max(high) as high,
       min(low) as low,
       last(close, time) as close
FROM stock_prices
WHERE symbol = '000001.SZ'
GROUP BY day;

InfluxDB​

from influxdb_client import InfluxDBClient

client = InfluxDBClient(url="http://localhost:8086", token="your-token")
write_api = client.write_api()

# 写入数据
from influxdb_client import Point
point = Point("stock_price") \
    .tag("symbol", "000001.SZ") \
    .field("close", 10.5) \
    .time("2024-01-01T09:30:00Z")
write_api.write(bucket="stocks", record=point)

实用工具库​

技术指标​

import pandas_ta as ta

# 计算指标
df.ta.sma(length=20, append=True)      # 简单移动平均
df.ta.rsi(length=14, append=True)       # RSI
df.ta.macd(append=True)                 # MACD
df.ta.bbands(length=20, append=True)    # 布林带

# 批量计算所有指标
df.ta.strategy(ta.AllStrategy)

财务计算​

import numpy_financial as npf

# 计算IRR
cash_flows = [-1000, 300, 400, 400, 300]
irr = npf.irr(cash_flows)

# 计算NPV
npv = npf.npv(0.1, cash_flows)  # 折现率10%

选择建议​

初学者​

  1. AKShare + Jupyter 快速开始
  2. Backtrader 学习回测
  3. 聚宽/米筐 了解平台化流程

进阶开发者​

  1. vn.py 实盘交易
  2. VectorBT 大规模研究
  3. 自研系统 完全控制

机构级别​

  1. Wind/iFinD 专业数据
  2. C++核心 + Python接口
  3. KDB+ 高性能时序数据库

延伸阅读​