{"id":2254,"date":"2025-05-07T07:42:00","date_gmt":"2025-05-06T23:42:00","guid":{"rendered":"https:\/\/blog.laoyulaoyu.top\/?p=2254"},"modified":"2025-04-28T16:10:15","modified_gmt":"2025-04-28T08:10:15","slug":"%e9%87%8f%e5%8c%96%e5%9b%a2%e9%98%9f%e7%a7%81%e8%97%8fpython%e5%ba%93%e5%88%86%e4%ba%ab%ef%bc%8c%e8%ae%a9%e4%bd%a03%e5%88%86%e9%92%9f%e5%ad%a6%e4%bc%9a%e6%9c%ba%e6%9e%84%e7%ba%a7%e8%82%a1%e7%a5%a8","status":"publish","type":"post","link":"https:\/\/laoyulaoyu.com\/index.php\/2025\/05\/07\/%e9%87%8f%e5%8c%96%e5%9b%a2%e9%98%9f%e7%a7%81%e8%97%8fpython%e5%ba%93%e5%88%86%e4%ba%ab%ef%bc%8c%e8%ae%a9%e4%bd%a03%e5%88%86%e9%92%9f%e5%ad%a6%e4%bc%9a%e6%9c%ba%e6%9e%84%e7%ba%a7%e8%82%a1%e7%a5%a8\/","title":{"rendered":"\u91cf\u5316\u56e2\u961f\u79c1\u85cfPython\u5e93\u5206\u4eab\uff0c\u8ba9\u4f603\u5206\u949f\u5b66\u4f1a\u673a\u6784\u7ea7\u80a1\u7968\u7ec4\u5408\u4f18\u5316"},"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>\u548c\u5927\u5bb6\u5206\u4eab\u4e00\u4e2a\u91cf\u5316\u56e2\u961f\u7ecf\u5e38\u4f1a\u7528\u7684PyPortfolioOpt\u5e93\uff0c\u4ed6\u4eec\u4f1a\u7528\u8fd9\u4e2a\u5e93\u7ed3\u5408AI\u91cf\u5316\u6280\u672f\uff0c\u7528\u6765\u8dd1\u673a\u6784\u7ea7\u6700\u4f18\u7ec4\u5408<\/strong><\/mark>\u3002\u63a5\u4e0b\u6765\u6211\u8ba9\u6211\u7528\u51e0\u5206\u949f\uff0c\u8ba9\u4f60\u8f7b\u677e\u638c\u63e1\u98ce\u9669\u4e0e\u6536\u76ca\u7684\u5e73\u8861\uff0c\u5b66\u4f1a\u98ce\u9669\u8c03\u4f18\u3002<\/pre>\n<\/blockquote>\n\n\n\n
\u4e00\u3001\u4e3a\u4ec0\u4e48\u8981\u505a\u80a1\u7968\u7ec4\u5408\u4f18\u5316\uff1f<\/strong><\/h2>\n\n\n\n
\u4f20\u7edf\u7092\u80a1\u5c31\u95ed\u773c\u6254\u98de\u9556\uff0c\u800c\u91cf\u5316\u80a1\u7968\u6295\u8d44\u7ec4\u5408\u5c31\u662f\u7ed9\u98de\u9556\u88c5\u7784\u51c6\u5668\u3002<\/p>\n\n\n\n
\u4e3e\u4e2a\u4f8b\u5b50\uff1a2023\u5e74\u7f8e\u80a1”\u79d1\u6280\u4e03\u5de8\u5934”\u5e73\u5747\u6da8105%\uff0c\u4f46\u5982\u679c\u4f60\u5168\u4ed3\u82f1\u4f1f\u8fbe\uff0c\u4eca\u5e741\u6708\u56de\u8c03\u65f6\u4f60\u53ef\u80fd\u54ed\u6655\u5728\u5395\u6240\u2014\u2014\u8fd9\u5c31\u662f\u4e0d\u61c2\u98ce\u9669\u5206\u6563\u7684\u4ee3\u4ef7\u3002<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\u4e70\u80a1\u7968\u4e0d\u53ea\u662f\u51ed\u611f\u89c9\uff0c\u8fd8\u8981\u770b\u201c\u56de\u62a5\/\u98ce\u9669\u201d\u6bd4\u3002<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\u800c\u80a1\u7968\u6295\u8d44\u7ec4\u5408\u6709\u5f88\u591a\u79cd\u7b56\u7565\uff0c\u4eca\u5929\u6211\u4eec\u91cd\u70b9\u8bb2\u4e00\u8bb2\u57fa\u4e8ePython\u7684PyPortfolioOpt\u5e93\u7684AI\u91cf\u5316\u667a\u80fd\u4f18\u5316<\/strong>\u6a21\u5f0f\u3002<\/p>\n\n\n\n
\u4e8c\u3001\u5b9e\u6218\u89e3\u6790\u539f\u7406\uff1a\u7b49\u6743 vs \u4f18\u5316<\/strong><\/h2>\n\n\n\n
\u76f8\u6bd4\u7b80\u5355\u7684\u7b49\u6743\u914d\u7f6e\uff0c\u4f18\u5316\u540e\u7ec4\u5408\u5f80\u5f80\u80fd\u5728\u76f8\u540c\u98ce\u9669\u6c34\u5e73\u4e0b\u83b7\u53d6\u66f4\u9ad8\u56de\u62a5\uff0c\u4e14\u901a\u8fc7\u91cf\u5316\u6846\u67b6\u964d\u4f4e\u4e3b\u89c2\u5224\u65ad\u5931\u8bef\u3002\u63a5\u4e0b\u6765\u6211\u7ed9\u5927\u5bb6\u505a \u7b49\u6743 vs \u4f18\u5316<\/strong> \u7684\u5177\u4f53\u5206\u6790\u3002<\/p>\n\n\n\n
2.1 “\u96e8\u9732\u5747\u6cbe”\u6d3e\uff1a\u7b49\u6743\u5e73\u5747\u5206\u914d<\/strong><\/h3>\n\n\n\n
\u6838\u5fc3\u73a9\u6cd5<\/strong>\uff1a\u6bcf\u53ea\u80a1\u7968\u90fd\u5206\u5230\u540c\u6837\u7684\u94b1\uff0c\u5c31\u50cf\u7ed9\u6bcf\u4e2a\u5b69\u5b50\u53d1\u76f8\u540c\u6570\u91cf\u7684\u7cd6\u679c\u3002<\/p>\n\n\n\n
\u4f18\u52bf<\/strong>\uff1a<\/p>\n\n\n\n
    \n
  • \u64cd\u4f5c\u7b80\u5355\uff0c\u624b\u673a1\u5206\u949f\u641e\u5b9a\uff1b<\/li>\n\n\n\n
  • \u907f\u514d\u88ab\u67d0\u53ea\u66b4\u96f7\u80a1\u5751\u60e8\uff08\u6bd4\u5982\u53bb\u5e74\u67d0\u5927\u836f\u5382\u4e24\u5929\u8dcc60%\uff09\u3002<\/li>\n<\/ul>\n\n\n\n
    \u7f3a\u70b9<\/strong>\uff1a<\/p>\n\n\n\n
    \u53ef\u80fd\u9519\u8fc7\u8d85\u7ea7\u725b\u80a1\uff082017\u5e74\u4f60\u5e73\u5747\u4e70BAT\uff0c\u817e\u8baf\u7684\u6da8\u5e45\u4f1a\u88ab\u53e6\u5916\u4e24\u5bb6\u62d6\u7d2f\uff09<\/p>\n\n\n\n
    \u5b9e\u6218\u8868\u73b0<\/strong>\uff1a\u7814\u7a76\u663e\u793a\uff1a2010-2020\u5e74\u6807\u666e500\u7b49\u6743\u6307\u6570\u5e74\u5316\u6536\u76ca8.7%\uff0c\u6bd4\u5e02\u503c\u52a0\u6743\u6307\u6570\u9ad81.2%\uff0c\u4f462021\u5e74\u5143\u5b87\u5b99\u6ce1\u6cab\u7834\u706d\u65f6\u591a\u8dcc\u4e865%\u3002<\/p>\n\n\n\n
    2.2 “\u7cbe\u51c6\u6253\u51fb”\u6d3e\uff1aAI\u667a\u80fd\u4f18\u5316<\/strong><\/h3>\n\n\n\n
    \u5728\u73b0\u4ee3\u6295\u8d44\u4e2d\uff0c\u7ec4\u5408\u4f18\u5316\u901a\u8fc7 \u5747\u503c\u2014\u65b9\u5dee\u7406\u8bba<\/strong>\uff08MPT\uff09\u548c\u6709\u6548\u524d\u6cbf<\/strong>\u65b9\u6cd5\uff0c\u5e2e\u52a9\u6295\u8d44\u8005\u5728\u201c\u9884\u671f\u6536\u76ca\u201d\u4e0e\u201c\u627f\u62c5\u98ce\u9669\u201d\u4e4b\u95f4\u627e\u5230\u6700\u4f18\u5e73\u8861\u3002<\/p>\n\n\n\n
    \u9ed1\u79d1\u6280\u539f\u7406<\/strong>\uff1a<\/p>\n\n\n\n
      \n
    • \u901a\u8fc7”\u667a\u80fd\u5929\u5e73”\u7b97\u6cd5\uff08\u539fMPT\u7406\u8bba\uff09\uff1b<\/li>\n\n\n\n
    • \u81ea\u52a8\u7b5b\u9009”\u8d5a\u94b1\u80fd\u529b\/\u6ce2\u52a8\u98ce\u9669”\u6bd4\u6700\u9ad8\u7684\u80a1\u7968\uff1b<\/li>\n\n\n\n
    • \u91cd\u70b9\u914d\u7f6e2-3\u53ea\u5f3a\u52bf\u80a1\u5f62\u6210\u8d5a\u94b1\u7a81\u51fb\u961f\u3002<\/li>\n<\/ul>\n\n\n\n
      2.3 \u5b9e\u6218\u6548\u679c\u5bf9\u6bd4<\/strong><\/h3>\n\n\n\n
      \u4ee52023\u5e74\u7f8e\u80a1\u4e03\u5de8\u5934\u4e3a\u4f8b\uff1a<\/p>\n\n\n\n
      \u6307\u6807<\/th>\u5e73\u5747\u5206\u914d\u6d3e<\/th>AI\u4f18\u5316\u6d3e<\/th>\u53d8\u5316\u8d8b\u52bf<\/th><\/tr><\/thead>
      \u5e74\u5316\u6536\u76ca<\/td>63.05%<\/td>105.56%<\/td>\u2191\ud83d\ude8042.51%<\/td><\/tr>
      \u6ce2\u52a8\u7387<\/td>23.98%<\/td>37.05%<\/td>\u2191\ud83d\udca313.07%<\/td><\/tr>
      \u6027\u4ef7\u6bd4\u6307\u6807<\/td>2.63<\/td>2.85<\/td>\u2191\ud83d\udd3c8.4%<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n
      \u91cd\u70b9\u89e3\u6790<\/strong>\uff1a<\/p>\n\n\n\n
        \n
      • \u6536\u76ca\u66b4\u6da8\u5bc6\u7801\uff1aAI\u7ed9\u82f1\u4f1f\u8fbe(NVDA)\u914d\u4e8628%\u4ed3\u4f4d\uff0c\u5143\u5b87\u5b99(META)\u7ed9\u4e8622%\uff1b<\/li>\n\n\n\n
      • \u4ee3\u4ef7\uff1a\u9047\u5230\u677f\u5757\u56de\u8c03\u65f6\uff08\u59822024\u5e741\u6708AI\u80a1\u5927\u8dcc\uff09\uff0c\u7ec4\u5408\u6ce2\u52a8\u582a\u6bd4\u5750\u8fc7\u5c71\u8f66\uff1b<\/li>\n\n\n\n
      • \u9002\u5408\u4eba\u7fa4\uff1a\u80fd\u627f\u53d720%\u4ee5\u4e0a\u6ce2\u52a8\u7684\u6fc0\u8fdb\u6295\u8d44\u8005\u3002<\/li>\n<\/ul>\n\n\n\n
        \u5177\u4f53\u8868\u73b0\u5dee\u5f02<\/strong>\uff1a<\/p>\n\n\n\n
        \u7ec4\u5408<\/th>\u9884\u671f\u5e74\u5316\u6536\u76ca<\/th>\u5e74\u5316\u6ce2\u52a8\u7387<\/th>\u590f\u666e\u6bd4<\/th><\/tr><\/thead>
        \u7b49\u6743\u91cd<\/td>63.05%<\/td>23.98%<\/td>2.63<\/td><\/tr>
        \u4f18\u5316\uff08Max Sharpe\uff09<\/td>105.56%<\/td>37.05%<\/td>2.85<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n
        \u6536\u76ca\u63d0\u5347<\/strong>\uff1a+42.51 \u767e\u5206\u70b9\uff0c\u6e90\u4e8e\u4f18\u5316\u5bf9 NVDA\u3001META \u7684\u91cd\u5ea6\u914d\u7f6e\uff1b<\/p>\n\n\n\n
        \u98ce\u9669\u653e\u5927<\/strong>\uff1a\u6ce2\u52a8\u7387\u589e\u52a0 13.07 \u767e\u5206\u70b9\uff0c\u56e0\u8d44\u4ea7\u9009\u62e9\u66f4\u96c6\u4e2d\uff1b<\/p>\n\n\n\n
        \u98ce\u9669\u8c03\u6574\u540e\u8868\u73b0<\/strong>\uff1a\u590f\u666e\u6bd4\u63d0\u5347 ~8.4%\uff0c\u8bc1\u660e\u6bcf\u5355\u4f4d\u98ce\u9669\u5e26\u6765\u66f4\u591a\u8d85\u989d\u6536\u76ca\u3002<\/p>\n\n\n\n
        2.4 \u539f\u7406\u5c0f\u7ed3<\/strong><\/h3>\n\n\n\n
          \n
        • \u98ce\u9669\u2013\u6536\u76ca\u6743\u8861<\/strong>\uff1a\u9ad8\u56de\u62a5\u6807\u7684\u5f80\u5f80\u4f34\u968f\u9ad8\u6ce2\u52a8\uff0c\u4f18\u5316\u5728\u6536\u76ca\u63d0\u5347\u7684\u540c\u65f6\uff0c\u4e5f\u9700\u627f\u53d7\u6ce2\u52a8\u589e\u52a0\uff1b<\/li>\n\n\n\n
        • \u5206\u6563 vs \u96c6\u4e2d<\/strong>\uff1a\u4f18\u5316\u7ec4\u5408\u5c06\u653e\u5f03\u90e8\u5206\u5206\u6563\u4f18\u52bf\u4ee5\u8d62\u53d6\u66f4\u9ad8\u56de\u62a5\uff1b<\/li>\n\n\n\n
        • \u6a21\u578b\u5047\u8bbe\u5c40\u9650<\/strong>\uff1a\u8f93\u5165\uff08\u6536\u76ca\u3001\u534f\u65b9\u5dee\uff09\u4f30\u8ba1\u8bef\u5dee\u4f1a\u5f71\u54cd\u7ed3\u679c\uff0c\u9700\u7a33\u5065\u6027\u68c0\u9a8c\u4e0e\u7ea6\u675f\u7ba1\u7406\u3002<\/li>\n<\/ul>\n\n\n\n
          \u4e09\u3001\u624b\u628a\u624b\u6559\u5b66<\/strong><\/h2>\n\n\n\n
          \u4e0b\u9762\u5206\u4e09\u6b65\uff0c\u4e3a\u5927\u5bb6\u9010\u4e00\u5256\u6790\u5982\u4f55\u5728\u7528 Python \u642d\u5efa\u4e00\u4e2a\u5b8c\u6574\u7684\u201c\u6293\u6570\u636e\u2192\u7b97\u6307\u6807\u2192\u8c03\u6a21\u578b\u201d\u6d41\u7a0b\u3002<\/p>\n\n\n\n
          3.1 \u6570\u636e\u83b7\u53d6\u4e0e\u6e05\u6d17<\/strong><\/h3>\n\n\n\n
            \n
          • \u5de5\u5177\u9009\u578b<\/strong>\uff1a\u91c7\u7528 yfinance<\/code>\uff0c\u652f\u6301\u591a\u53ea\u80a1\u7968\u4e00\u6b21\u6027\u62c9\u53d6\u5386\u53f2\u6570\u636e\u3002<\/li>\n\n\n\n
          • \u5173\u952e\u601d\u8def<\/strong>\uff1a\u62c9\u53d6 \u8c03\u6574\u540e\u6536\u76d8\u4ef7<\/strong>\uff08\u8003\u8651\u5206\u7ea2\u62c6\u80a1\uff09\uff0c\u5e76\u4f7f\u7528\u524d\u5411\u586b\u5145 fillna<\/code> \u5904\u7406\u5076\u53d1\u7f3a\u5931\u3002 <\/li>\n<\/ul>\n\n\n\n
            import yfinance as yf\n\ntickers = [\"AAPL\",\"MSFT\",\"AMZN\",\"GOOGL\",\"META\",\"NVDA\"]\ndata = yf.download(\n    tickers,\n    start=\"2023-01-01\",\n    end=\"2024-12-31\",\n    auto_adjust=True     # \u81ea\u52a8\u8c03\u6574\u4ef7\u683c\n)[\"Close\"].fillna(method=\"ffill\")<\/code><\/pre>\n\n\n\n
            auto_adjust=True<\/code> \u53ef\u81ea\u52a8\u5c06\u62c6\u80a1\u3001\u5206\u7ea2\u4f53\u73b0\u5230\u4ef7\u683c\u4e2d\uff0c\u907f\u514d\u540e\u671f\u624b\u52a8\u8c03\u6574\u3002<\/p>\n\n\n\n
            3.2 \u8ba1\u7b97\u9884\u671f\u6536\u76ca\u4e0e\u98ce\u9669<\/strong><\/h3>\n\n\n\n
              \n
            • \u9884\u671f\u6536\u76ca<\/strong>\uff1a\u7528\u5386\u53f2\u65e5\u6536\u76ca\u7684\u5e74\u5316\u5e73\u5747\u4f30\u7b97\uff0c\u7b49\u540c\u4e8e \u671f\u671b\u6536\u76ca = \u65e5\u6536\u76ca\u7387\u5747\u503c \u00d7 252<\/code>\u3002<\/li>\n\n\n\n
            • \u534f\u65b9\u5dee\u77e9\u9635<\/strong>\uff1a\u6837\u672c\u534f\u65b9\u5dee\u523b\u753b\u5404\u80a1\u4e4b\u95f4\u7684\u8054\u52a8\u6027\uff0c\u98ce\u9669\u6a21\u578b\u53ef\u9009\u57fa\u7840\u6837\u672c\u3001\u6307\u6570\u52a0\u6743\u6216\u6536\u7f29\u7b49\u65b9\u6cd5\u3002 <\/li>\n<\/ul>\n\n\n\n
              from pypfopt import expected_returns, risk_models\n\n# \u5386\u53f2\u5747\u503c\u6cd5\uff1a\u5e74\u5316\u671f\u671b\u6536\u76ca\nmu = expected_returns.mean_historical_return(data)\n\n# \u6837\u672c\u534f\u65b9\u5dee\uff1a\u8ba1\u7b97\u65e5\u6536\u76ca\u534f\u65b9\u5dee\u540e\u5e74\u5316\nS  = risk_models.sample_cov(data)<\/code><\/pre>\n\n\n\n
              3.3 \u6784\u5efa\u6a21\u578b\u5e76\u751f\u6210\u6700\u4f18\u7ec4\u5408<\/strong><\/h3>\n\n\n\n
                \n
              • \u6709\u6548\u524d\u6cbf\u6846\u67b6<\/strong>\uff1a\u57fa\u4e8e\u5747\u503c\u2014\u65b9\u5dee\u76ee\u6807\uff0c\u8c03\u7528 EfficientFrontier<\/code> \u7c7b\u81ea\u52a8\u6c42\u89e3\u6700\u4f18\u7ec4\u5408\u3002<\/li>\n\n\n\n
              • \u6700\u5927\u5316\u590f\u666e\u6bd4<\/strong>\uff1a\u5373\u5728\u5254\u9664\u65e0\u98ce\u9669\u6536\u76ca\u540e\uff0c\u4f7f\u201c\u8d85\u989d\u6536\u76ca\/\u6ce2\u52a8\u201d\u6700\u5927\u3002<\/li>\n<\/ul>\n\n\n\n
                from pypfopt import EfficientFrontier\n\nef = EfficientFrontier(mu, S)\nraw_weights = ef.max_sharpe()      # \u6700\u5927\u590f\u666e\u6bd4\nclean_weights = ef.clean_weights() # \u6e05\u9664\u5fae\u5c0f\u6743\u91cd<\/code><\/pre>\n\n\n\n
                3.4 \u4f18\u5316\u9a71\u52a8\u5185\u56e0<\/strong><\/h3>\n\n\n\n
                \u5728\u672c\u6848\u4f8b\u4e2d\uff0c\u7b97\u6cd5\u7ed9 NVDA \u914d\u7f6e\u4e86 56.78%\u3001\u7ed9 META \u914d\u7f6e\u4e86 36.02%\uff0c\u800c\u57fa\u672c\u5ffd\u7565\u4e86 AAPL\u3001AMZN\u3001GOOGL\u3001MSFT\u3002<\/p>\n\n\n\n
                \u5747\u503c\u2014\u65b9\u5dee\u6743\u91cd\u8f6c\u79fb<\/strong>
                \u5747\u503c\u2014\u65b9\u5dee\u4f18\u5316\u4f1a\u81ea\u52a8\u5c06\u8d44\u91d1\u5f80\u201c\u56de\u62a5\/\u98ce\u9669\u201d\u6bd4\u66f4\u9ad8\u7684\u8d44\u4ea7\u503e\u659c\u3002<\/p>\n\n\n\n
                \"\"<\/figure>\n\n\n\n
                \u96c6\u4e2d\u5ea6\u4e0e\u5206\u6563\u5ea6\u7684\u53d6\u820d<\/strong><\/p>\n\n\n\n