{"id":1650,"date":"2024-11-21T07:41:00","date_gmt":"2024-11-20T23:41:00","guid":{"rendered":"https:\/\/blog.laoyulaoyu.top\/?p=1650"},"modified":"2024-10-27T20:44:10","modified_gmt":"2024-10-27T12:44:10","slug":"%e5%bf%85%e7%9c%8b%ef%bc%81%e8%82%a1%e7%a5%a8%e9%85%8d%e5%af%b9%e4%ba%a4%e6%98%93%e7%bb%9f%e8%ae%a1%e5%a5%97%e5%88%a9%e5%ae%9e%e6%88%98%e7%bb%88%e6%9e%81%e6%8c%87%e5%8d%97","status":"publish","type":"post","link":"https:\/\/www.laoyulaoyu.com\/index.php\/2024\/11\/21\/%e5%bf%85%e7%9c%8b%ef%bc%81%e8%82%a1%e7%a5%a8%e9%85%8d%e5%af%b9%e4%ba%a4%e6%98%93%e7%bb%9f%e8%ae%a1%e5%a5%97%e5%88%a9%e5%ae%9e%e6%88%98%e7%bb%88%e6%9e%81%e6%8c%87%e5%8d%97\/","title":{"rendered":"\u5fc5\u770b\uff01\u80a1\u7968\u914d\u5bf9\u4ea4\u6613\u7edf\u8ba1\u5957\u5229\u5b9e\u6218\u7ec8\u6781\u6307\u5357"},"content":{"rendered":"\n

\u4f5c\u8005\uff1a\u8001\u4f59\u635e\u9c7c<\/a><\/p>\n\n\n\n

\u539f\u521b\u4e0d\u6613\uff0c\u8f6c\u8f7d\u8bf7\u6807\u660e\u51fa\u5904\u53ca\u539f\u4f5c\u8005\u3002<\/mark><\/strong><\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n
\n
\u5199\u5728\u524d\u9762\u7684\u8bdd\uff1a<\/strong>\u6211\u66fe\u64b0\u5199\u8fc7\u4e24\u7bc7\u6709\u5173\u914d\u5bf9\u4ea4\u6613\u7edf\u8ba1\u5957\u5229\u7684\u6587\u7ae0\uff0c\u5176\u4e2d\u4e00\u7bc7\u300a\u4e3a\u4f60\u63ed\u5f00\u80a1\u7968\u914d\u5bf9\u4ea4\u6613\u7edf\u8ba1\u5957\u5229\u7684\u5b9e\u73b0\u300b\u53d1\u8868\u540e\uff0c\u6709\u8bfb\u8005\u4e0e\u6211\u4ea4\u6d41\uff0c\u8868\u793a\u5176\u76f8\u5173\u7406\u8bba\u548c\u7a0b\u5e8f\u6846\u67b6\u90fd\u5df2\u638c\u63e1\uff0c\u4f46\u5728\u5b9e\u9645\u64cd\u4f5c\u65b9\u9762\u4ecd\u7f3a\u4e4f\u4e86\u89e3\u3002\u4eca\u5929\uff0c\u6211\u5728\u524d\u6587\u7684\u57fa\u7840\u4e0a\u7eed\u5199\u4e86\u4e00\u7bc7\uff0c\u4ecb\u7ecd\u5982\u4f55\u5bf9\u82f9\u679c\uff08AAPL\uff09\u548c\u5fae\u8f6f\uff08MSFT\uff09\u80a1\u7968\u8fdb\u884c\u914d\u5bf9\u4ea4\u6613\u7684\u5b9e\u64cd<\/mark>\uff0c\u5e76\u901a\u8fc7\u56de\u6d4b\u5411\u5927\u5bb6\u5c55\u793a\u8be5\u7b56\u7565\u7684\u8868\u73b0\u3002<\/pre>\n<\/blockquote>\n\n\n\n
\u4e00\u3001\u6211\u4eec\u8981\u505a\u4ec0\u4e48\uff1f <\/em><\/strong><\/h3>\n\n\n\n
\u7d22\u7f57\u65af\uff08George Soros<\/em>\uff09\u8bf4\uff1a”\u91cd\u8981\u7684\u4e0d\u662f\u4f60\u662f\u5bf9\u662f\u9519\uff0c\u800c\u662f\u4f60\u5bf9\u7684\u65f6\u5019\u8d5a\u4e86\u591a\u5c11\u94b1\uff0c\u9519\u7684\u65f6\u5019\u4e8f\u4e86\u591a\u5c11\u94b1”<\/strong>\u3002\u201cIt\u2019s not whether you\u2019re right or wrong that\u2019s important, but how much money you make when you\u2019re right and how much you lose when you\u2019re wrong.\u201d <\/em><\/p>\n\n\n\n
\u5728\u80a1\u5e02\u4e2d\uff0c\u77ed\u77ed\u51e0\u79d2\u5c31\u80fd\u8ba9\u8d44\u91d1\u5927\u5e45\u589e\u51cf\uff0c\u6240\u4ee5\u4ea4\u6613\u8005\u603b\u662f\u5728\u7b56\u7565\u91cc\u5bfb\u627e\u98ce\u9669\u4e0e\u6536\u76ca\u7684\u5e73\u8861\u70b9\u3002\u914d\u5bf9\u4ea4\u6613\u5c31\u662f\u5176\u4e2d\u4e00\u79cd\u3002\u4f7f\u7528\u8fd9\u79cd\u5e02\u573a\u4e2d\u6027\u7b56\u7565\uff0c\u4ea4\u6613\u8005\u4e0d\u5fc5\u5173\u5fc3\u7edd\u5bf9\u4ef7\u683c\u53d8\u5316\uff0c\u53ea\u7528\u5173\u6ce8\u4e24\u79cd\u76f8\u5173\u5de5\u5177\u4e4b\u95f4\u7684\u4ef7\u683c\u5dee\u3002<\/p>\n\n\n\n
\u5728\u672c\u6587\u4e2d\uff0c\u6211\u4eec\u5c06\u4ee5\u4e24\u5bb6\u7a33\u5b9a\u589e\u957f\u7684\u5168\u7403\u6280\u672f\u9886\u519b\u4f01\u4e1a\u4e3a\u4f8b\uff0c\u5c55\u793a\u5982\u4f55\u8fd0\u7528\u914d\u5bf9\u4ea4\u6613\uff1a\u82f9\u679c\uff08AAPL\uff09\u548c\u5fae\u8f6f\uff08MSFT\uff09\u3002\u6211\u5c06\u9610\u8ff0\u5982\u4f55\u5b9e\u65bd\u8fd9\u6837\u4e00\u79cd\u7edf\u8ba1\u5957\u5229\u7b56\u7565\uff0c\u6db5\u76d6\u534f\u6574\u5206\u6790\u3001\u5bf9\u51b2\u6bd4\u7387\u786e\u5b9a\u548c\u4ef7\u5dee\u6784\u5efa\u7b49\u6b65\u9aa4\u3002\u6b64\u5916\uff0c\u6211\u8fd8\u5c06\u4ecb\u7ecd\u5982\u4f55\u5bf9\u8be5\u7b56\u7565\u8fdb\u884c\u56de\u6eaf\u6d4b\u8bd5\uff0c\u5e76\u4f9d\u636e\u5305\u62ec\u6536\u76ca\u3001\u6ce2\u52a8\u7387\u3001\u590f\u666e\u6bd4\u7387\u548c\u6700\u5927\u56de\u64a4\u7b49\u56e0\u7d20\u6765\u8bc4\u4f30\u5176\u7ed3\u679c\u3002\u6700\u7ec8\uff0c\u60a8\u5c06\u4e86\u89e3\u914d\u5bf9\u4ea4\u6613\u7684\u57fa\u672c\u6d41\u7a0b\uff0c\u4ee5\u53ca\u8fd9\u79cd\u7b56\u7565\u5728\u5b9e\u9645\u5e02\u573a\u73af\u5883\u4e2d\u7684\u51fa\u8272\u8868\u73b0\u3002<\/p>\n\n\n\n
\u4e8c\u3001\u5bfc\u5165\u5e93\u548c\u6570\u636e\u51c6\u5907<\/strong><\/h3>\n\n\n\n
\u9996\u5148\uff0c\u6211\u4eec\u9700\u8981\u5bfc\u5165\u51e0\u4e2a Python \u5e93\uff0c\u5b83\u4eec\u5c06\u5e2e\u52a9\u6211\u4eec\u5206\u6790\u6570\u636e\u5e76\u8fd0\u884c\u914d\u5bf9\u4ea4\u6613\u7b56\u7565\u3002<\/p>\n\n\n\n
import pandas as pd\nimport numpy as np\nimport yfinance as yf\nfrom sklearn.linear_model import LinearRegression\nfrom statsmodels.tsa.stattools import coint\nimport matplotlib.pyplot as plt\nimport seaborn as sns<\/code><\/code><\/pre>\n\n\n\n
\u8fd9\u4e9b\u5e93\u5bf9\u4e8e\u5904\u7406\u6570\u636e\u3001\u8fd0\u884c\u7edf\u8ba1\u6d4b\u8bd5\u548c\u7ed8\u5236\u7ed3\u679c\u81f3\u5173\u91cd\u8981\u3002\u96c5\u864e\u8d22\u7ecf\uff08yfinance\uff09\u63d0\u4f9b\u4e86\u82f9\u679c\u548c\u5fae\u8f6f\u7684\u5386\u53f2\u80a1\u7968\u6570\u636e\u3002Pandas \u6709\u52a9\u4e8e\u6570\u636e\u5904\u7406\uff0c\u800c NumPy \u5219\u652f\u6301\u6570\u503c\u8ba1\u7b97\u3002\u8bbf\u95ee\u5e72\u51c0\u7684\u5386\u53f2\u6570\u636e\u662f\u4efb\u4f55\u6210\u529f\u4ea4\u6613\u7b56\u7565\u7684\u57fa\u7840\u3002\u5b83\u4f7f\u6211\u4eec\u80fd\u591f\u51c6\u786e\u5730\u68c0\u67e5\u4ef7\u683c\u8d70\u52bf\u3001\u6267\u884c\u6d4b\u8bd5\u548c\u56de\u6d4b\u7b56\u7565\u3002<\/p>\n\n\n\n
\u4e09\u3001\u4e0b\u8f7d\u76ee\u6807\u80a1\u7968\u4ef7\u683c\u6570\u636e<\/strong><\/h3>\n\n\n\n
\u73b0\u5728\uff0c\u6211\u4eec\u5c06\u4e0b\u8f7d\u82f9\u679c\u516c\u53f8\uff08AAPL\uff09\u548c\u5fae\u8f6f\u516c\u53f8\uff08MSFT\uff09\u4ece 2015 \u5e74 1 \u6708\u5230 2023 \u5e74 10 \u6708\u7684\u5386\u53f2\u6570\u636e\u3002<\/p>\n\n\n\n
ticker1 = 'AAPL'
ticker2 = 'MSFT'
start_date = '2015-01-01'
end_date = '2024-10-01'
data1 = yf.download(ticker1, start=start_date, end=end_date)
data2 = yf.download(ticker2, start=start_date, end=end_date)<\/pre>\n\n\n\n
\u8be5\u4ee3\u7801\u80fd\u591f\u83b7\u53d6\u4e24\u5bb6\u516c\u53f8\u7684\u8c03\u6574\u540e\u6536\u76d8\u4ef7\u3002\u8c03\u6574\u540e\u7684\u6536\u76d8\u4ef7\u6781\u4e3a\u91cd\u8981\uff0c\u56e0\u4e3a\u5b83\u6db5\u76d6\u4e86\u6240\u6709\u7684\u516c\u53f8\u884c\u4e3a\uff0c\u4f8b\u5982\u5206\u7ea2\u548c\u80a1\u7968\u5206\u5272\uff0c\u4ece\u800c\u771f\u5b9e\u5730\u4f53\u73b0\u4e86\u80a1\u7968\u7684\u4ef7\u503c\u3002<\/p>\n\n\n\n
\u62e5\u6709\u7cbe\u786e\u7684\u3001\u7ecf\u8fc7\u8c03\u6574\u7684\u4ef7\u683c\u6570\u636e\uff0c\u80fd\u591f\u786e\u4fdd\u6211\u4eec\u7684\u5206\u6790\u5efa\u7acb\u5728\u73b0\u5b9e\u7684\u3001\u53cd\u6620\u5e02\u573a\u7684\u4ef7\u503c\u57fa\u7840\u4e4b\u4e0a\u3002\u8fd9\u5bf9\u4e8e\u5206\u6790\u957f\u671f\u8d8b\u52bf\u4ee5\u53ca\u9884\u6d4b\u672a\u6765\u8d70\u52bf\u5177\u6709\u81f3\u5173\u91cd\u8981\u7684\u610f\u4e49\u3002<\/p>\n\n\n\n
\u56db\u3001\u4ef7\u683c\u5e8f\u5217\u53ef\u89c6\u5316<\/strong><\/h3>\n\n\n\n
\u6211\u4eec\u6765\u76f4\u89c2\u5730\u770b\u5230\u82f9\u679c\u548c\u5fae\u8f6f\u7684\u4ef7\u683c\u968f\u65f6\u95f4\u7684\u53d8\u5316\u60c5\u51b5\u3002<\/p>\n\n\n\n
plt.figure(figsize=(14, 7))\nplt.plot(df[ticker1], label=ticker1)\nplt.plot(df[ticker2], label=ticker2)\nplt.title('Price Series of AAPL and MSFT')\nplt.xlabel('Date')\nplt.ylabel('Adjusted Close Price')\nplt.legend()\nplt.show()<\/code><\/code><\/pre>\n\n\n\n
\u4e0b\u56fe\u663e\u793a AAPL \u548c MSFT \u7684\u4ef7\u683c\u5e8f\u5217\u53ca\u5176\u8c03\u6574\u540e\u7684\u6536\u76d8\u4ef7\u3002<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\u5f53\u6211\u4eec\u7ed8\u5236 AAPL \u548c MSFT \u7684\u8c03\u6574\u540e\u6536\u76d8\u4ef7\u65f6\uff0c\u5c31\u80fd\u770b\u51fa\u8fd9\u4e24\u53ea\u80a1\u7968\u81ea 2015 \u5e74\u4ee5\u6765\u7684\u8d70\u52bf\u662f\u591a\u4e48\u76f8\u8fd1\u3002\u4ef7\u683c\u5e8f\u5217\u7684\u53ef\u89c6\u5316\u8ba9\u6211\u4eec\u80fd\u591f\u8fc5\u901f\u5224\u65ad\u8fd9\u4e24\u53ea\u80a1\u7968\u5728\u4e00\u6bb5\u65f6\u95f4\u5185\u7684\u8d70\u52bf\u662f\u5426\u76f8\u4f3c\u3002\u8fd9\u5728\u914d\u5bf9\u4ea4\u6613\u4e2d\u975e\u5e38\u5173\u952e\uff0c\u56e0\u4e3a\u4e24\u79cd\u8d44\u4ea7\u4e4b\u95f4\u7684\u5f3a\u76f8\u5173\u6027\u662f\u8be5\u7b56\u7565\u53d1\u6325\u4f5c\u7528\u7684\u5fc5\u8981\u6761\u4ef6\u3002<\/p>\n\n\n\n
\u4e94\u3001\u534f\u6574\u68c0\u9a8c<\/strong><\/h3>\n\n\n\n
\u8fd9\u53ef\u4ee5\u5e2e\u52a9\u6211\u4eec\u786e\u5b9a\u4e24\u4e2a\u80a1\u7968\u4ef7\u683c\u662f\u5426\u5177\u6709\u7edf\u8ba1\u610f\u4e49\u4e0a\u7684\u957f\u671f\u5173\u7cfb\u3002<\/p>\n\n\n\n
score, pvalue, _ = coint(df[ticker1], df[ticker2])\nprint(f'Cointegration test p-value: {pvalue:.4f}')<\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
p \u503c\u4e3a 0.1100\uff0c\u8868\u660e AAPL \u548c MSFT \u4e4b\u95f4\u5b58\u5728\u5fae\u5f31\u7684\u534f\u6574\u5173\u7cfb\u3002<\/p>\n\n\n\n
\u534f\u6574\u68c0\u9a8c\u68c0\u9a8c AAPL \u548c MSFT \u4e4b\u95f4\u662f\u5426\u5b58\u5728\u957f\u671f\u5747\u8861\u5173\u7cfb\u3002\u5982\u679c p \u503c\u4f4e\u4e8e 0.05\uff0c\u5219\u8868\u660e\u5b58\u5728\u8f83\u5f3a\u7684\u534f\u6574\u5173\u7cfb\uff0c\u5373\u8fd9\u4e24\u53ea\u80a1\u7968\u7684\u4ef7\u683c\u4f1a\u968f\u7740\u65f6\u95f4\u7684\u63a8\u79fb\u800c\u5171\u540c\u53d8\u52a8\u3002\u867d\u7136 p \u503c\u9ad8\u4e8e\u6807\u51c6\u4e34\u754c\u503c\uff0c\u4f46\u7b56\u7565\u4ecd\u53ef\u7ee7\u7eed\u3002\u5373\u4f7f\u534f\u6574\u5173\u7cfb\u4e0d\u5b8c\u7f8e\uff0c\u914d\u5bf9\u4ea4\u6613\u4e5f\u80fd\u53d1\u6325\u4f5c\u7528\uff0c\u53ea\u662f\u9700\u8981\u66f4\u4ed4\u7ec6\u5730\u76d1\u63a7\u4ef7\u5dee\u53d8\u52a8\uff0c\u5e76\u53ef\u80fd\u6d89\u53ca\u66f4\u9ad8\u7684\u98ce\u9669\u3002<\/p>\n\n\n\n
\u516d\u3001\u5bf9\u51b2\u6bd4\u7387\u7684\u7ebf\u6027\u56de\u5f52<\/strong><\/h3>\n\n\n\n
\u4e0b\u4e00\u6b65\u662f\u8ba1\u7b97\u5bf9\u51b2\u6bd4\u7387\uff0c\u8fd9\u6709\u52a9\u4e8e\u5e73\u8861\u4e24\u4e2a\u5934\u5bf8\u3002<\/p>\n\n\n\n
X = df[ticker2].values.reshape(-1, 1)\ny = df[ticker1].values\nmodel = LinearRegression()\nmodel.fit(X, y)\nhedge_ratio = model.coef_[0]\nprint(f'Hedge Ratio: {hedge_ratio:.4f}')<\/code><\/code><\/pre>\n\n\n\n
\u8fd9\u4e2a\u7ebf\u6027\u56de\u5f52\u6a21\u578b\u53cd\u6620\u4e86\u82f9\u679c\u548c\u5fae\u8f6f\u4ef7\u683c\u4e4b\u95f4\u7684\u5173\u7cfb\u3002\u7531\u6b64\u5f97\u51fa\u7684\u5bf9\u51b2\u6bd4\u7387\uff080.5558\uff09\u544a\u8bc9\u6211\u4eec\uff0c\u5728\u6211\u4eec\u7684\u914d\u5bf9\u4ea4\u6613\u4e2d\uff0c\u6bcf\u53ea\u80a1\u7968\u6211\u4eec\u5e94\u8be5\u6301\u6709\u591a\u5c11\u3002<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\u5957\u671f\u4fdd\u503c\u6bd4\u7387\u5bf9\u4e8e\u5efa\u7acb\u5e02\u573a\u4e2d\u6027\u7684\u6295\u8d44\u7ec4\u5408\u81f3\u5173\u91cd\u8981\u3002\u901a\u8fc7\u6b63\u786e\u6301\u6709\u6bcf\u53ea\u80a1\u7968\u7684\u6570\u91cf\uff0c\u6211\u4eec\u53ef\u4ee5\u786e\u4fdd\u6211\u4eec\u7684\u7b56\u7565\u4ece\u5b83\u4eec\u7684\u76f8\u5bf9\u4ef7\u683c\u53d8\u52a8\u4e2d\u83b7\u5229\uff0c\u800c\u4e0d\u662f\u4ece\u6574\u4e2a\u5e02\u573a\u7684\u8d8b\u52bf\u4e2d\u83b7\u5229\u3002<\/p>\n\n\n\n
\u4e03\u3001\u8ba1\u7b97\u548c\u7ed8\u5236\u4ef7\u5dee\u56fe<\/strong><\/h3>\n\n\n\n
\u4ef7\u5dee\u662f AAPL \u548c MSFT \u4ef7\u683c\u4e4b\u95f4\u7684\u5dee\u989d\uff0c\u6309\u5bf9\u51b2\u6bd4\u7387\u8fdb\u884c\u8c03\u6574\u3002<\/p>\n\n\n\n
df['Spread'] = df[ticker1] - hedge_ratio * df[ticker2]<\/code><\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\u4ef7\u5dee\u663e\u793a\u4e86\u4e24\u53ea\u80a1\u7968\u4ef7\u683c\u76f8\u5bf9\u79fb\u52a8\u7684\u8ddd\u79bb\u3002\u8fd9\u662f\u6211\u4eec\u8fdb\u884c\u4ea4\u6613\u7684\u5173\u952e\u503c\u3002\u914d\u5bf9\u4ea4\u6613\u7b56\u7565\u4f9d\u8d56\u4e8e\u5747\u503c\u56de\u5f52\uff0c\u8fd9\u610f\u5473\u7740\u6211\u4eec\u9884\u671f\u4ef7\u5dee\u6700\u7ec8\u4f1a\u56de\u5f52\u5230\u5386\u53f2\u5e73\u5747\u6c34\u5e73\u3002\u786e\u5b9a\u4ef7\u5dee\u8fc7\u5bbd\u6216\u8fc7\u7a84\u65f6\uff0c\u6211\u4eec\u5c31\u53ef\u4ee5\u8fdb\u884c\u4ea4\u6613\u3002<\/p>\n\n\n\n
\u516b\u3001Z \u503c\u8ba1\u7b97\u4e0e\u4e70\u5356\u4fe1\u53f7\u751f\u6210<\/strong><\/h3>\n\n\n\n
\u4e3a\u4e86\u91cf\u5316\u5dee\u503c\u4e0e\u5e73\u5747\u503c\u4e4b\u95f4\u7684\u8ddd\u79bb\uff0c\u6211\u4eec\u8ba1\u7b97 Z \u503c\u3002<\/p>\n\n\n\n
df['Z_Score'] = (df['Spread'] - df['Spread_Mean']) \/ df['Spread_STD']<\/code><\/code><\/pre>\n\n\n\n
Z \u503c\u8861\u91cf\u5f53\u524d\u4ef7\u5dee\u4e0e\u5176\u5e73\u5747\u503c\u7684\u6807\u51c6\u5dee\u3002\u8fd9\u6709\u52a9\u4e8e\u6211\u4eec\u8bc6\u522b\u6781\u7aef\u504f\u5dee\uff0c\u56e0\u4e3a\u6781\u7aef\u504f\u5dee\u9884\u793a\u7740\u6f5c\u5728\u7684\u4ea4\u6613\u673a\u4f1a\u3002<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\u4f7f\u7528 Z \u503c\u65f6\uff0c\u6211\u4eec\u53ef\u4ee5\u8bbe\u7f6e\u9608\u503c\u6765\u89e6\u53d1\u4e70\u5165\u548c\u5356\u51fa\u4fe1\u53f7\u3002\u8f83\u5927\u7684\u6b63 Z \u503c\u6216\u8d1f Z \u503c\u8868\u660e\u4ef7\u5dee\u8fdc\u79bb\u5e73\u5747\u503c\uff0c\u8868\u660e\u4ef7\u5dee\u53ef\u80fd\u4f1a\u56de\u8c03\u3002\u63a5\u4e0b\u6765\u6211\u4eec\u5229\u7528 Z \u503c\u751f\u6210\u505a\u591a\u6216\u505a\u7a7a\u7684\u4fe1\u53f7\u3002<\/p>\n\n\n\n
df['Long_Signal'] = np.where(df['Z_Score'] <= -entry_threshold, 1, 0)\ndf['Short_Signal'] = np.where(df['Z_Score'] >= entry_threshold, -1, 0)\ndf['Positions'] = df['Long_Signal'] + df['Short_Signal']<\/code><\/code><\/pre>\n\n\n\n
\u5f53 Z \u503c\u4f4e\u4e8e\u67d0\u4e00\u9608\u503c\u65f6\uff0c\u5c31\u4f1a\u53d1\u51fa\u4e70\u5165\uff08\u505a\u591a\uff09\u4fe1\u53f7\uff0c\u800c\u5f53 Z \u503c\u8d85\u8fc7\u6b63\u9608\u503c\u65f6\uff0c\u5c31\u4f1a\u53d1\u51fa\u5356\u51fa\uff08\u505a\u7a7a\uff09\u4fe1\u53f7\u3002\u8fd9\u4e9b\u4fe1\u53f7\u6784\u6210\u4e86\u8be5\u7b56\u7565\u7684\u652f\u67f1\u3002\u5b83\u5141\u8bb8\u6211\u4eec\u5728\u4ef7\u5dee\u4e25\u91cd\u504f\u79bb\u5747\u503c\u65f6\u8fdb\u5165\u4ea4\u6613\uff0c\u5e76\u5047\u5b9a\u4ef7\u5dee\u6700\u7ec8\u4f1a\u56de\u8c03\u3002<\/p>\n\n\n\n
\u4e5d\u3001\u7b56\u7565\u56de\u6eaf\u6d4b\u8bd5<\/strong><\/h3>\n\n\n\n
\u4e0b\u4e00\u6b65\u8ba9\u6211\u4eec\u8ba1\u7b97\u4e0b\u7b56\u7565\u7684\u7d2f\u8ba1\u56de\u62a5\u3002<\/p>\n\n\n\n
df['Strategy_Return'] = df['Positions'] * (df['Return_AAPL'] - hedge_ratio * df['Return_MSFT'])\ndf['Cumulative_Strategy_Return'] = (1 + df['Strategy_Return'].fillna(0)).cumprod() - 1<\/code><\/code><\/pre>\n\n\n\n
\u6211\u4eec\u901a\u8fc7\u5e94\u7528\u6211\u4eec\u7684\u4ea4\u6613\u4fe1\u53f7\u5bf9\u7b56\u7565\u8fdb\u884c\u56de\u6eaf\u6d4b\u8bd5\uff0c\u5e76\u8ba1\u7b97\u968f\u7740\u65f6\u95f4\u7684\u63a8\u79fb\u4ea4\u6613\u7684\u76c8\u5229\u60c5\u51b5\u3002<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\u901a\u8fc7\u56de\u6eaf\u6d4b\u8bd5\u8ba1\u7b97\u7684 2015-2024 \u5e74 AAPL \u548c MSFT \u80a1\u7968\u7d2f\u8ba1\u7b56\u7565\u56de\u62a5\u7387\uff08\u672a\u6263\u9664\u4ea4\u6613\u6210\u672c\uff09\uff0c\u4e0a\u56fe\u5411\u6211\u4eec\u5c55\u793a\u4e86\u8be5\u7b56\u7565\u7684\u5386\u53f2\u8868\u73b0\uff0c\u8ba9\u6211\u4eec\u6df1\u5165\u4e86\u89e3\u5176\u6f5c\u5728\u76c8\u5229\u80fd\u529b\u548c\u7a33\u5065\u6027\u3002<\/p>\n\n\n\n
\u5341\u3001\u4ea4\u6613\u6210\u672c\u4e0e\u51c0\u6218\u7565\u56de\u62a5<\/strong><\/h3>\n\n\n\n
\u8ba1\u5165\u4ea4\u6613\u6210\u672c\u540e\uff0c\u6211\u4eec\u5bf9\u6536\u76ca\u8fdb\u884c\u8c03\u6574\u3002<\/p>\n\n\n\n
df['Strategy_Return_Net'] = df['Strategy_Return'] - df['Transaction_Costs']\ndf['Cumulative_Strategy_Return_Net'] = (1 + df['Strategy_Return_Net'].fillna(0)).cumprod() - 1<\/code><\/code><\/pre>\n\n\n\n
\u8fd9\u4e00\u6b65\u6263\u9664\u4e86\u4ea4\u6613\u6210\u672c\uff0c\u4ee5\u4fbf\u66f4\u771f\u5b9e\u5730\u53cd\u6620\u6218\u7565\u7684\u76c8\u5229\u80fd\u529b\u3002\u4ea4\u6613\u6210\u672c\u4f1a\u5927\u5927\u964d\u4f4e\u5229\u6da6\uff0c\u5c24\u5176\u662f\u5728\u9ad8\u9891\u4ea4\u6613\u7b56\u7565\u4e2d\u3002\u5c06\u8fd9\u4e9b\u6210\u672c\u8ba1\u7b97\u5728\u5185\uff0c\u53ef\u4ee5\u786e\u4fdd\u6211\u4eec\u6e05\u695a\u5730\u4e86\u89e3\u7b56\u7565\u7684\u53ef\u884c\u6027\u3002<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n

\u5341\u4e00\u3001\u7ee9\u6548\u6307\u6807<\/strong><\/h3>\n\n\n\n
\u6700\u540e\uff0c\u6211\u4eec\u8ba1\u7b97\u5173\u952e\u6027\u80fd\u6307\u6807\u3002<\/p>\n\n\n\n
print(f'Annualized Return: {annualized_return:.2%}')\nprint(f'Annualized Volatility: {annualized_volatility:.2%}')\nprint(f'Sharpe Ratio: {sharpe_ratio:.2f}')\nprint(f'Maximum Drawdown: {max_drawdown:.2%}')<\/code><\/code><\/pre>\n\n\n\n
\u8ba1\u7b97\u5173\u952e\u7ee9\u6548\u6307\u6807\uff0c\u5305\u62ec\u5e74\u5316\u56de\u62a5\u7387\u3001\u6ce2\u52a8\u7387\u3001\u590f\u666e\u6bd4\u7387\u548c\u6700\u5927\u7f29\u6c34\u7387\uff0c\u4ee5\u8bc4\u4f30\u6218\u7565\u7684\u6709\u6548\u6027\u3002<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\u7ed3\u679c\u5982\u56fe\uff1a<\/strong><\/p>\n\n\n\n