一、核心环境配置

1. 基础Python环境

# 推荐使用 conda 创建独立环境
conda create -n quant python=3.9
conda activate quant

# 或者使用 venv
python -m venv quant_env
source quant_env/bin/activate  # Linux/Mac
# 或 quant_env\Scripts\activate  # Windows

2. 核心包安装

# 基础数据分析
pip install numpy pandas scipy statsmodels scikit-learn

# 量化金融专用
pip install ta-lib  # 技术指标库（需要单独安装依赖）
pip install empyrical  # 绩效分析
pip install pyfolio  # 组合分析
pip install riskfolio-lib  # 组合优化

# 替代ta-lib的简便方案
pip install ta  # 纯Python实现的技术指标




二、回测框架选择

1. Backtrader - 功能全面

import backtrader as bt

class MyStrategy(bt.Strategy):
    def __init__(self):
        self.sma = bt.indicators.SimpleMovingAverage(self.data, period=20)
    
    def next(self):
        if self.data.close[0] > self.sma[0]:
            self.buy()
        elif self.data.close[0] < self.sma[0]:
            self.sell()

# 安装
pip install backtrader


2. Zipline - 专业量化

from zipline import run_algorithm
from zipline.api import order, symbol

def initialize(context):
    context.asset = symbol('AAPL')

def handle_data(context, data):
    order(context.asset, 10)

# 安装较复杂，推荐使用conda
conda install -c conda-forge zipline


3. VectorBT - 向量化快速回测

import vectorbt as vbt

price = vbt.YFData.download('AAPL').get('Close')
fast_ma = vbt.MA.run(price, 10)
slow_ma = vbt.MA.run(price, 20)

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

pf = vbt.Portfolio.from_signals(price, entries, exits)
pf.stats()

# 安装
pip install vectorbt


4. Qlib - 微软开发的AI量化平台


from qlib.contrib.strategy import TopkDropoutStrategy
from qlib.contrib.evaluate import backtest

# 安装
pip install pyqlib




三、因子挖掘工作流

1. 数据获取模块
# 数据源配置
import pandas as pd
import yfinance as yf
import akshare as ak  # 国内数据
import tushare as ts  # 替代选择
import baostock as bs  # 免费数据源

# 安装
pip install yfinance akshare tushare baostock

2. 因子计算示例

import numpy as np
import pandas as pd

class FactorGenerator:
    def __init__(self, data):
        self.data = data
    
    def calculate_momentum(self, period=20):
        """动量因子"""
        return self.data['close'].pct_change(period)
    
    def calculate_volatility(self, window=20):
        """波动率因子"""
        return self.data['close'].pct_change().rolling(window).std()
    
    def calculate_rsi(self, period=14):
        """RSI因子"""
        delta = self.data['close'].diff()
        gain = (delta.where(delta > 0, 0)).rolling(window=period).mean()
        loss = (-delta.where(delta < 0, 0)).rolling(window=period).mean()
        rs = gain / loss
        return 100 - (100 / (1 + rs))


3. 因子检验模块

from alphalens import performance, plotting
import alphalens as al

def analyze_factor(factor_data, prices):
    """
    因子分析
    """
    # 因子数据预处理
    factor_data = al.utils.get_clean_factor_and_forward_returns(
        factor_data, 
        prices,
        quantiles=5,
        periods=(1, 5, 10)
    )
    
    # IC分析
    ic = al.performance.factor_information_coefficient(factor_data)
    
    # 分组收益
    mean_return_by_q, std_err_by_q = al.performance.mean_return_by_quantile(factor_data)
    
    return ic, mean_return_by_q

# 安装
pip install alphalens



四、完整项目结构示例

quant_project/
├── config/
│   ├── __init__.py
│   └── settings.py        # 配置文件
├── data/
│   ├── fetcher.py         # 数据获取
│   ├── processor.py       # 数据处理
│   └── storage.py         # 数据存储
├── factors/
│   ├── __init__.py
│   ├── technical.py       # 技术因子
│   ├── fundamental.py     # 基本面因子
│   └── custom.py          # 自定义因子
├── backtest/
│   ├── engine.py          # 回测引擎
│   ├── strategies.py      # 策略
│   └── analyzer.py        # 绩效分析
├── utils/
│   ├── helpers.py
│   └── metrics.py         # 评估指标
└── main.py




五、推荐开发环境
1. Jupyter Lab - 交互式研究

pip install jupyterlab
# 量化相关扩展
pip install jupyterlab-dash

2. VS Code配置


{
    "python.pythonPath": "quant_env/bin/python",
    "python.analysis.extraPaths": ["./quant_project"],
    "jupyter.notebookFileRoot": "${workspaceFolder}"
}

3. Docker环境（可选）


FROM python:3.9-slim
WORKDIR /app
COPY requirements.txt .
RUN pip install -r requirements.txt
COPY . .
CMD ["python", "main.py"]



六、快速开始的模板代码

# requirements.txt
numpy>=1.21.0
pandas>=1.3.0
backtrader==1.9.76.123
vectorbt==0.25.0
yfinance==0.1.70
empyrical==0.5.5
matplotlib>=3.4.0
seaborn>=0.11.0
jupyterlab>=3.0.0



# quick_start.py
import pandas as pd
import numpy as np
import vectorbt as vbt
import yfinance as yf
import matplotlib.pyplot as plt

# 1. 获取数据
data = yf.download('AAPL', start='2020-01-01', end='2023-01-01')

# 2. 计算简单因子
price = data['Close']
returns = price.pct_change()
sma_short = price.rolling(10).mean()
sma_long = price.rolling(30).mean()

# 3. 生成交易信号
entries = sma_short > sma_long
exits = sma_short < sma_long

# 4. 回测
pf = vbt.Portfolio.from_signals(
    price, 
    entries, 
    exits,
    fees=0.001,  # 交易费用
    freq='1D'
)

# 5. 分析结果
print(pf.stats())
pf.plot().show()




七、进阶工具推荐

机器学习集成
lightgbm / xgboost - 梯度提升树
scikit-learn - 传统机器学习
tensorflow / pytorch - 深度学习

优化工具
optuna - 超参数优化
hyperopt - 分布式超参数优化

风险分析
pyfolio - 组合绩效分析
arch - 波动率建模

八、注意事项
数据质量：确保数据清洗和复权处理
过拟合：使用交叉验证、避免未来函数
交易成本：考虑佣金、滑点、冲击成本
样本外测试：严格区分训练集和测试集