Finance \u2014 \u5168\u5e02\u573a\u91d1\u878d\u6570\u636e\u667a\u80fd\u5206\u6790\u5e73\u53f0 | \u591a\u6e90\u6570\u636e\u878d\u5408\u30fb\u56e0\u5b50\u6316\u6398\u30fb\u7edf\u8ba1\u5efa\u6a21\u30fb\u53ef\u89c6\u5316\u6d1e\u5bdf\u30fb\u91cf\u5316\u8d4b\u80fd\u4e00\u4f53\u5316\u5de5\u4f5c\u53f0<\/strong><\/p>\n\n\n\n \u4e00\u7ad9\u5f0f\u8986\u76d6\u80a1\u7968\u3001\u671f\u8d27\u3001\u57fa\u91d1\u3001\u503a\u5238\u3001\u6307\u6570\u3001\u5b8f\u89c2\u7ecf\u6d4e<\/strong>\u7684\u4e13\u4e1a\u91d1\u878d\u6570\u636e\u5206\u6790\u5f15\u64ce\uff0c\u96c6\u6570\u636e\u91c7\u96c6\u3001\u6e05\u6d17\u3001\u56e0\u5b50\u8ba1\u7b97\u3001\u7edf\u8ba1\u5206\u6790\u3001\u53ef\u89c6\u5316\u3001\u6570\u636e\u5efa\u6a21\u3001\u62a5\u8868\u5bfc\u51fa\u4e8e\u4e00\u4f53\u3002\u6df1\u5ea6\u6253\u901a\u524d\u6587 plaintext<\/p>\n\n\n\n txt<\/p>\n\n\n\n python<\/p>\n\n\n\n \u8fd0\u884c<\/p>\n\n\n\n python<\/p>\n\n\n\n \u8fd0\u884c<\/p>\n\n\n\n python<\/p>\n\n\n\n \u8fd0\u884c<\/p>\n\n\n\n python<\/p>\n\n\n\n \u8fd0\u884c<\/p>\n\n\n\n python<\/p>\n\n\n\n \u8fd0\u884c<\/p>\n\n\n\n python<\/p>\n\n\n\n \u8fd0\u884c<\/p>\n\n\n\n python<\/p>\n\n\n\n \u8fd0\u884c<\/p>\n\n\n\n python<\/p>\n\n\n\n \u8fd0\u884c<\/p>\n\n\n\n python<\/p>\n\n\n\n \u8fd0\u884c<\/p>\n\n\n\n python<\/p>\n\n\n\n \u8fd0\u884c<\/p>\n\n\n\n bash<\/p>\n\n\n\n \u8fd0\u884c<\/p>\n\n\n\n plaintext<\/p>\n\n\n\n plaintext<\/p>\n\n\n\n plaintext<\/p>\n\n\n\n \u4e00\u3001\u72ec\u7acb\u6807\u9898 & \u5dee\u5f02\u5316\u5b9a\u4f4d Finance \u2014 \u5168\u5e02\u573a\u91d1\u878d\u6570\u636e\u667a\u80fd\u5206\u6790\u5e73\u53f0 | \u591a\u6e90\u6570\u636e\u878d\u5408\u30fb\u56e0\u5b50\u6316 […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[],"class_list":["post-301","post","type-post","status-publish","format-standard","hentry","category-mianfeiercikaifalongxiaskill"],"_links":{"self":[{"href":"https:\/\/www.longxiapro.com\/index.php\/wp-json\/wp\/v2\/posts\/301","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.longxiapro.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.longxiapro.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.longxiapro.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.longxiapro.com\/index.php\/wp-json\/wp\/v2\/comments?post=301"}],"version-history":[{"count":1,"href":"https:\/\/www.longxiapro.com\/index.php\/wp-json\/wp\/v2\/posts\/301\/revisions"}],"predecessor-version":[{"id":302,"href":"https:\/\/www.longxiapro.com\/index.php\/wp-json\/wp\/v2\/posts\/301\/revisions\/302"}],"wp:attachment":[{"href":"https:\/\/www.longxiapro.com\/index.php\/wp-json\/wp\/v2\/media?parent=301"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.longxiapro.com\/index.php\/wp-json\/wp\/v2\/categories?post=301"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.longxiapro.com\/index.php\/wp-json\/wp\/v2\/tags?post=301"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}\u6838\u5fc3\u5b9a\u4f4d<\/h3>\n\n\n\n
Backtest Expert<\/code>\u3001Stock Monitor<\/code>\u3001Baidu Web Search<\/code>\u3001WeChat MP CN<\/code> \u7b49\u5de5\u5177\uff0c\u5b9e\u73b0\u884c\u60c5\u6570\u636e\u2192\u8206\u60c5\u6570\u636e\u2192\u91cf\u5316\u56e0\u5b50\u2192\u7b56\u7565\u56de\u6d4b\u2192\u98ce\u9669\u76d1\u63a7<\/strong>\u5168\u94fe\u8def\u95ed\u73af\uff0c\u517c\u987e\u6295\u7814\u5206\u6790\u3001\u91cf\u5316\u5f00\u53d1\u3001\u98ce\u63a7\u590d\u76d8\u3001\u884c\u60c5\u7814\u5224\u7b49\u573a\u666f\u3002<\/p>\n\n\n\n\u72ec\u5bb6\u6539\u9020\u7279\u8272\uff08\u533a\u522b\u4e8e\u666e\u901a\u6570\u636e\u5206\u6790\u5de5\u5177\uff09<\/h3>\n\n\n\n
\n
Memory Tiering<\/code>\uff0c\u9ad8\u9891\u884c\u60c5\u3001\u5b9e\u65f6\u56e0\u5b50\u5e38\u9a7b\u9ad8\u901f\u5185\u5b58\uff0c\u5386\u53f2\u8d22\u62a5\u3001\u5e74\u5ea6\u5b8f\u89c2\u6570\u636e\u4e0b\u6c89\u51b7\u5b58\u50a8\uff0c\u5343\u4e07\u7ea7\u6570\u636e\u8ba1\u7b97\u63d0\u901f 50%+\u3002<\/li>\n\n\n\nBacktest Expert<\/code> \u7b56\u7565\u56e0\u5b50\u3001Stock Monitor<\/code> \u76d1\u63a7\u6807\u7684\u3001Remotion \u52a8\u6001\u6570\u636e\u89c6\u9891\u3001\u793e\u5a92\u8c03\u5ea6\u5de5\u5177\u81ea\u52a8\u5206\u53d1\u7814\u62a5\u3002<\/li>\n\n\n\n\u4e8c\u3001\u4f20\u7edf\u91d1\u878d\u5206\u6790\u5de5\u5177\u75db\u70b9<\/h3>\n\n\n\n
\n
\u4e09\u3001\u6838\u5fc3\u529f\u80fd\u6a21\u5757<\/h3>\n\n\n\n
1. \u591a\u6e90\u6570\u636e\u63a5\u5165\u6a21\u5757<\/h3>\n\n\n\n
\n
2. \u6570\u636e\u6e05\u6d17 & \u6807\u51c6\u5316\u6a21\u5757<\/h3>\n\n\n\n
\n
3. \u91d1\u878d\u56e0\u5b50\u8ba1\u7b97\u6a21\u5757<\/h3>\n\n\n\n
\n
4. \u4e13\u4e1a\u7edf\u8ba1\u5206\u6790\u6a21\u5757<\/h3>\n\n\n\n
\n
5. \u53ef\u89c6\u5316\u5206\u6790\u6a21\u5757<\/h3>\n\n\n\n
\n
6. \u98ce\u9669\u76d1\u63a7 & \u5f02\u52a8\u9884\u8b66<\/h3>\n\n\n\n
\n
7. \u62a5\u8868 & \u5bfc\u51fa\u6a21\u5757<\/h3>\n\n\n\n
\n
8. \u751f\u6001\u5bf9\u63a5\u6a21\u5757<\/h3>\n\n\n\n
\n
Backtest Expert<\/code> \u8f93\u51fa\u56e0\u5b50\u6570\u636e\u3001\u6807\u7684\u6c60\uff0c\u7528\u4e8e\u7b56\u7565\u56de\u6d4b<\/li>\n\n\n\nStock Monitor<\/code> \u63a8\u9001\u4f18\u9009\u6807\u7684\u3001\u5f02\u52a8\u6807\u7684\uff0c\u5f00\u542f\u5b9e\u65f6\u76d1\u63a7<\/li>\n\n\n\nRemotion<\/code> \u5c06\u5206\u6790\u56fe\u8868\u751f\u6210\u52a8\u6001\u8bb2\u89e3\u89c6\u9891<\/li>\n\n\n\nSocial Media Scheduler<\/code> \u81ea\u52a8\u5206\u53d1\u7814\u62a5\u3001\u5206\u6790\u5185\u5bb9<\/li>\n<\/ul>\n\n\n\n\u56db\u3001\u9879\u76ee\u76ee\u5f55\u7ed3\u6784<\/h3>\n\n\n\n
finance\/\n\u251c\u2500\u2500 config\/ # \u5168\u5c40\u914d\u7f6e\u3001\u6570\u636e\u6e90\u3001\u544a\u8b66\u89c4\u5219\u3001\u7f13\u5b58\u7b56\u7565\n\u251c\u2500\u2500 core\/ # \u6570\u636e\u5185\u6838\u3001\u6570\u636e\u6e05\u6d17\u3001\u590d\u6743\u8ba1\u7b97\u3001\u65f6\u95f4\u5bf9\u9f50\n\u251c\u2500\u2500 data_source\/ # \u591a\u6570\u636e\u6e90\u5bf9\u63a5\uff08Tushare\/Akshare\/\u722c\u866b\uff09\n\u251c\u2500\u2500 factor\/ # \u91d1\u878d\u56e0\u5b50\u5e93\u3001\u81ea\u5b9a\u4e49\u56e0\u5b50\u5f15\u64ce\u3001\u56e0\u5b50\u68c0\u9a8c\n\u251c\u2500\u2500 statistics\/ # \u7edf\u8ba1\u5206\u6790\u3001\u56de\u5f52\u3001\u76f8\u5173\u6027\u3001\u5206\u5e03\u8ba1\u7b97\n\u251c\u2500\u2500 visual\/ # \u53ef\u89c6\u5316\u56fe\u8868\u3001\u7ed8\u56fe\u7ec4\u4ef6\n\u251c\u2500\u2500 risk_warn\/ # \u98ce\u9669\u8bc6\u522b\u3001\u5f02\u52a8\u68c0\u6d4b\u3001\u544a\u8b66\u63a8\u9001\n\u251c\u2500\u2500 report\/ # \u62a5\u8868\u6a21\u677f\u3001\u751f\u6210\u5f15\u64ce\u3001\u5bfc\u51fa\u5de5\u5177\n\u251c\u2500\u2500 api\/ # FastAPI\u63a5\u53e3\u3001\u5bf9\u5916\u670d\u52a1\u3001OpenClaw\u5bf9\u63a5\n\u251c\u2500\u2500 task\/ # \u5b9a\u65f6\u4efb\u52a1\u3001\u6279\u91cf\u8ba1\u7b97\u3001\u4efb\u52a1\u961f\u5217\n\u251c\u2500\u2500 utils\/ # \u5de5\u5177\u51fd\u6570\u3001\u65e5\u5fd7\u3001\u5185\u5b58\u5206\u5c42\u9002\u914d\n\u251c\u2500\u2500 web\/ # \u524d\u7aef\u4ea4\u4e92\u5f0f\u5206\u6790\u770b\u677f\n\u251c\u2500\u2500 requirements.txt # \u9879\u76ee\u4f9d\u8d56\n\u2514\u2500\u2500 main.py # \u9879\u76ee\u5165\u53e3\n<\/code><\/pre>\n\n\n\n\u4e94\u3001\u5b8c\u6574\u4e8c\u6b21\u5f00\u53d1\u4ee3\u7801<\/h3>\n\n\n\n
1. requirements.txt<\/h3>\n\n\n\n
# \u6570\u636e\u83b7\u53d6\ntushare>=1.7.9\nakshare>=1.12.0\nrequests>=2.31.0\nbeautifulsoup4>=4.12.0\n\n# \u6570\u636e\u5904\u7406\nnumpy>=1.24.0\npandas>=2.0.0\nscipy>=1.10.0\n\n# \u7edf\u8ba1\u5efa\u6a21\nscikit-learn>=1.3.0\nstatsmodels>=0.14.0\n\n# \u53ef\u89c6\u5316\nmatplotlib>=3.7.0\nplotly>=5.17.0\nseaborn>=0.13.0\n\n# \u63a5\u53e3\u4e0e\u8c03\u5ea6\nfastapi>=0.104.0\nuvicorn>=0.24.0\napscheduler>=3.10.0\n\n# \u7f13\u5b58\u4e0e\u5de5\u5177\nredis>=5.0.0\nloguru>=0.7.2\npython-dotenv>=1.0.0\nopenpyxl>=3.1.0\njinja2>=3.1.0\n<\/code><\/pre>\n\n\n\n2. \u5168\u5c40\u914d\u7f6e config\/settings.py<\/h3>\n\n\n\n
import os\nfrom dotenv import load_dotenv\n\nload_dotenv()\n\n# \u6570\u636e\u6e90\u5bc6\u94a5\nTUSHARE_TOKEN = os.getenv(\"TUSHARE_TOKEN\", \"\")\n\n# \u6570\u636e\u5b58\u50a8\u8def\u5f84\nRAW_DATA_DIR = \".\/data\/raw\"\nCLEAN_DATA_DIR = \".\/data\/clean\"\nREPORT_DIR = \".\/report\/output\"\n\n# \u590d\u6743\u8bbe\u7f6e: qfq\u524d\u590d\u6743 \/ hfq\u540e\u590d\u6743 \/ none\u4e0d\u590d\u6743\nADJUST_TYPE = \"qfq\"\n\n# \u5185\u5b58\u5206\u5c42\u914d\u7f6e\uff08\u5bf9\u63a5Memory Tiering\uff09\nHOT_DATA_EXPIRE = 1800\nCOLD_DATA_COMPRESS = True\n\n# \u98ce\u9669\u544a\u8b66\u9608\u503c\nPRICE_CHANGE_WARN = 0.05 # \u6da8\u8dcc5%\u89e6\u53d1\u544a\u8b66\nNET_MONEY_WARN = 50000000 # \u5355\u65e5\u8d44\u91d1\u5f02\u52a85000\u4e07\n\n# \u63a5\u53e3\u914d\u7f6e\nAPI_HOST = \"0.0.0.0\"\nAPI_PORT = 8100\n\n# \u5b9a\u65f6\u4efb\u52a1\u65f6\u533a\nTIMEZONE = \"Asia\/Shanghai\"\n<\/code><\/pre>\n\n\n\n3. \u65e5\u5fd7\u5de5\u5177 utils\/logger.py<\/h3>\n\n\n\n
from loguru import logger\nimport os\nfrom config.settings import RAW_DATA_DIR\n\nlog_file = os.path.join(RAW_DATA_DIR, \"finance_run.log\")\n\nlogger.add(\n log_file,\n rotation=\"500 MB\",\n retention=\"30 days\",\n level=\"INFO\",\n encoding=\"utf-8\",\n enqueue=True\n)\n\ndef get_logger():\n return logger\n<\/code><\/pre>\n\n\n\n4. \u6570\u636e\u6e90\u5f15\u64ce data_source\/data_loader.py<\/h3>\n\n\n\n
import tushare\nimport pandas as pd\nfrom datetime import datetime\nfrom config.settings import TUSHARE_TOKEN, ADJUST_TYPE\nfrom utils.logger import get_logger\n\nlogger = get_logger()\ntushare.set_token(TUSHARE_TOKEN)\npro = tushare.pro_api()\n\nclass DataLoader:\n def __init__(self):\n self.cache = {}\n\n def get_daily_bar(self, ts_code, start_date, end_date):\n \"\"\"\u83b7\u53d6\u65e5\u7ebf\u884c\u60c5 + \u590d\u6743\u5904\u7406\"\"\"\n cache_key = f\"{ts_code}_{start_date}_{end_date}\"\n if cache_key in self.cache:\n return self.cache[cache_key]\n\n try:\n # \u539f\u59cb\u884c\u60c5\n df = pro.daily(\n ts_code=ts_code,\n start_date=start_date.replace(\"-\", \"\"),\n end_date=end_date.replace(\"-\", \"\")\n )\n if df.empty:\n logger.warning(f\"\u65e0\u884c\u60c5\u6570\u636e\uff1a{ts_code}\")\n return None\n\n # \u590d\u6743\u56e0\u5b50\n adj_factor = pro.adj_factor(\n ts_code=ts_code,\n start_date=start_date.replace(\"-\", \"\"),\n end_date=end_date.replace(\"-\", \"\")\n )\n df = df.merge(adj_factor, on=\"trade_date\", how=\"left\")\n df[\"adj_factor\"] = df[\"adj_factor\"].fillna(1)\n\n # \u590d\u6743\u8ba1\u7b97\n if ADJUST_TYPE == \"qfq\":\n df[\"open\"] = df[\"open\"] * df[\"adj_factor\"]\n df[\"high\"] = df[\"high\"] * df[\"adj_factor\"]\n df[\"low\"] = df[\"low\"] * df[\"low\"] * df[\"adj_factor\"]\n df[\"close\"] = df[\"close\"] * df[\"adj_factor\"]\n elif ADJUST_TYPE == \"hfq\":\n latest_factor = df[\"adj_factor\"].iloc[0]\n df[\"open\"] = df[\"open\"] * df[\"adj_factor\"] \/ latest_factor\n df[\"high\"] = df[\"high\"] * df[\"adj_factor\"] \/ latest_factor\n df[\"low\"] = df[\"low\"] * df[\"adj_factor\"] \/ latest_factor\n df[\"close\"] = df[\"close\"] * df[\"adj_factor\"] \/ latest_factor\n\n # \u65e5\u671f\u683c\u5f0f\u5316 & \u6392\u5e8f\n df[\"trade_date\"] = pd.to_datetime(df[\"trade_date\"])\n df.rename(columns={\"trade_date\": \"date\"}, inplace=True)\n df = df.sort_values(\"date\").reset_index(drop=True)\n df = df[df[\"vol\"] > 0]\n\n self.cache[cache_key] = df\n logger.info(f\"\u884c\u60c5\u52a0\u8f7d\u5b8c\u6210\uff1a{ts_code} \u5171{len(df)}\u6761\")\n return df\n except Exception as e:\n logger.error(f\"\u884c\u60c5\u83b7\u53d6\u5931\u8d25 {ts_code}\uff1a{str(e)}\")\n return None\n\n def get_money_flow(self, ts_code, start_date, end_date):\n \"\"\"\u83b7\u53d6\u4e2a\u80a1\u8d44\u91d1\u6d41\u5411\"\"\"\n try:\n df = pro.moneyflow(\n ts_code=ts_code,\n start_date=start_date.replace(\"-\", \"\"),\n end_date=end_date.replace(\"-\", \"\")\n )\n df[\"trade_date\"] = pd.to_datetime(df[\"trade_date\"])\n df.rename(columns={\"trade_date\": \"date\"}, inplace=True)\n return df.sort_values(\"date\").reset_index(drop=True)\n except Exception as e:\n logger.error(f\"\u8d44\u91d1\u6d41\u83b7\u53d6\u5931\u8d25\uff1a{str(e)}\")\n return None\n\n# \u5168\u5c40\u5b9e\u4f8b\ndata_loader = DataLoader()\n<\/code><\/pre>\n\n\n\n5. \u6570\u636e\u6e05\u6d17\u5185\u6838 core\/data_clean.py<\/h3>\n\n\n\n
import pandas as pd\nimport numpy as np\nfrom utils.logger import get_logger\n\nlogger = get_logger()\n\nclass DataCleaner:\n @staticmethod\n def fill_missing(df, method=\"ffill\"):\n \"\"\"\u7f3a\u5931\u503c\u586b\u5145\uff1a\u524d\u5411\/\u540e\u5411\/\u5747\u503c\u586b\u5145\"\"\"\n if method == \"ffill\":\n df = df.fillna(method=\"ffill\")\n elif method == \"bfill\":\n df = df.fillna(method=\"bfill\")\n elif method == \"mean\":\n df = df.fillna(df.mean())\n return df\n\n @staticmethod\n def filter_abnormal_price(df, threshold=0.2):\n \"\"\"\u8fc7\u6ee4\u5f02\u5e38\u4ef7\u683c\uff08\u5355\u65e5\u6da8\u8dcc\u5e45\u8d85\u9608\u503c\uff09\"\"\"\n df[\"pct_change\"] = df[\"close\"].pct_change()\n df = df[(df[\"pct_change\"] > -threshold) & (df[\"pct_change\"] < threshold)]\n df = df.drop(columns=[\"pct_change\"])\n return df.reset_index(drop=True)\n\n @staticmethod\n def standardize_column(df):\n \"\"\"\u5b57\u6bb5\u6807\u51c6\u5316\uff0c\u7edf\u4e00\u5206\u6790\u53e3\u5f84\"\"\"\n standard_cols = [\"date\", \"open\", \"high\", \"low\", \"close\", \"vol\", \"amount\"]\n exist_cols = [c for c in standard_cols if c in df.columns]\n return df[exist_cols]\n\n# \u5168\u5c40\u6e05\u6d17\u5b9e\u4f8b\ndata_cleaner = DataCleaner()\n<\/code><\/pre>\n\n\n\n6. \u91d1\u878d\u56e0\u5b50\u5e93 factor\/finance_factor.py<\/h3>\n\n\n\n
import pandas as pd\nimport numpy as np\n\nclass FinanceFactor:\n @staticmethod\n def calc_momentum(df, period=20):\n \"\"\"\u52a8\u91cf\u56e0\u5b50\uff1aN\u65e5\u6da8\u8dcc\u5e45\"\"\"\n df[f\"momentum_{period}\"] = df[\"close\"].pct_change(periods=period)\n return df\n\n @staticmethod\n def calc_volatility(df, period=20):\n \"\"\"\u6ce2\u52a8\u7387\u56e0\u5b50\"\"\"\n df[\"ret\"] = df[\"close\"].pct_change()\n df[f\"vol_{period}\"] = df[\"ret\"].rolling(window=period).std() * np.sqrt(250)\n df.drop(columns=[\"ret\"], inplace=True)\n return df\n\n @staticmethod\n def calc_turnover(df):\n \"\"\"\u6362\u624b\u7387\u56e0\u5b50\uff08\u7b80\u6613\uff09\"\"\"\n df[\"turnover\"] = df[\"vol\"] \/ df[\"vol\"].rolling(60).mean()\n return df\n\n @staticmethod\n def calc_all_factors(df):\n \"\"\"\u6279\u91cf\u8ba1\u7b97\u5e38\u7528\u56e0\u5b50\"\"\"\n df = FinanceFactor.calc_momentum(df, 10)\n df = FinanceFactor.calc_momentum(df, 20)\n df = FinanceFactor.calc_volatility(df, 20)\n df = FinanceFactor.calc_turnover(df)\n return df\n\n# \u5168\u5c40\u56e0\u5b50\u5b9e\u4f8b\nfactor_calc = FinanceFactor()\n<\/code><\/pre>\n\n\n\n7. \u7edf\u8ba1\u5206\u6790\u6a21\u5757 statistics\/analysis.py<\/h3>\n\n\n\n
import pandas as pd\nimport numpy as np\nfrom scipy.stats import pearsonr\nfrom sklearn.linear_model import LinearRegression\n\nclass FinanceAnalysis:\n @staticmethod\n def basic_stats(df):\n \"\"\"\u57fa\u7840\u6536\u76ca\u3001\u98ce\u9669\u7edf\u8ba1\"\"\"\n df[\"ret\"] = df[\"close\"].pct_change()\n total_ret = (df[\"close\"].iloc[-1] \/ df[\"close\"].iloc[0]) - 1\n annual_ret = (1 + total_ret) ** (250 \/ len(df)) - 1\n volatility = df[\"ret\"].std() * np.sqrt(250)\n\n # \u6700\u5927\u56de\u64a4\n cum_max = df[\"close\"].cummax()\n drawdown = (df[\"close\"] - cum_max) \/ cum_max\n max_dd = drawdown.min()\n\n return {\n \"total_return\": round(total_ret, 4),\n \"annual_return\": round(annual_ret, 4),\n \"volatility\": round(volatility, 4),\n \"max_drawdown\": round(max_dd, 4)\n }\n\n @staticmethod\n def corr_analysis(df1, df2, col=\"close\"):\n \"\"\"\u4e24\u7ec4\u6570\u636e\u76f8\u5173\u6027\u5206\u6790\"\"\"\n s1 = df1[col].pct_change().dropna()\n s2 = df2[col].pct_change().dropna()\n # \u5bf9\u9f50\u957f\u5ea6\n min_len = min(len(s1), len(s2))\n s1, s2 = s1[:min_len], s2[:min_len]\n corr, p_value = pearsonr(s1, s2)\n return {\"correlation\": round(corr, 4), \"p_value\": round(p_value, 4)}\n\n @staticmethod\n def rolling_regression(df, x_col, y_col, window=60):\n \"\"\"\u6eda\u52a8\u7ebf\u6027\u56de\u5f52\"\"\"\n df = df.copy().dropna()\n coef_list = []\n for i in range(window, len(df)):\n slice_df = df.iloc[i-window:i]\n X = slice_df[[x_col]].values\n y = slice_df[y_col].values\n model = LinearRegression()\n model.fit(X, y)\n coef_list.append(model.coef_[0])\n return {\"rolling_coef\": coef_list}\n\n# \u5206\u6790\u5b9e\u4f8b\nfinance_analyst = FinanceAnalysis()\n<\/code><\/pre>\n\n\n\n8. \u98ce\u9669\u5f02\u52a8\u544a\u8b66 risk_warn\/warning.py<\/h3>\n\n\n\n
from config.settings import PRICE_CHANGE_WARN, NET_MONEY_WARN\nfrom utils.logger import get_logger\n\nlogger = get_logger()\n\nclass RiskWarning:\n @staticmethod\n def check_price_warn(df):\n \"\"\"\u4ef7\u683c\u6da8\u8dcc\u5f02\u52a8\u544a\u8b66\"\"\"\n if len(df) < 2:\n return False, \"\"\n last_ret = df[\"close\"].pct_change().iloc[-1]\n if abs(last_ret) >= PRICE_CHANGE_WARN:\n msg = f\"\u4ef7\u683c\u5f02\u52a8\u544a\u8b66\uff0c\u5f53\u65e5\u6da8\u8dcc\u5e45\uff1a{last_ret:.2%}\"\n logger.warning(msg)\n return True, msg\n return False, \"\"\n\n @staticmethod\n def check_money_warn(money_df):\n \"\"\"\u8d44\u91d1\u6d41\u5f02\u52a8\u544a\u8b66\"\"\"\n if len(money_df) < 1:\n return False, \"\"\n last_net = money_df[\"net_amount\"].iloc[-1]\n if abs(last_net) >= NET_MONEY_WARN:\n msg = f\"\u8d44\u91d1\u5f02\u52a8\u544a\u8b66\uff0c\u5355\u65e5\u51c0\u989d\uff1a{last_net\/10000:.0f}\u4e07\u5143\"\n logger.warning(msg)\n return True, msg\n return False, \"\"\n\nwarn_engine = RiskWarning()\n<\/code><\/pre>\n\n\n\n9. \u62a5\u8868\u751f\u6210\u6a21\u5757 report\/report_generator.py<\/h3>\n\n\n\n
import pandas as pd\nfrom jinja2 import Environment, FileSystemLoader\nimport os\nfrom config.settings import REPORT_DIR\nfrom utils.logger import get_logger\n\nlogger = get_logger()\n\nclass ReportGenerator:\n def __init__(self):\n self.env = Environment(loader=FileSystemLoader(\".\/report\/template\"))\n\n def generate_html_report(self, data, stats, title=\"\u91d1\u878d\u5206\u6790\u62a5\u544a\"):\n \"\"\"\u751f\u6210HTML\u53ef\u89c6\u5316\u62a5\u544a\"\"\"\n template = self.env.get_template(\"base.html\")\n html = template.render(title=title, data=data.to_html(), stats=stats)\n save_path = os.path.join(REPORT_DIR, f\"{title}.html\")\n with open(save_path, \"w\", encoding=\"utf-8\") as f:\n f.write(html)\n logger.info(f\"HTML\u62a5\u544a\u5df2\u751f\u6210\uff1a{save_path}\")\n return save_path\n\n def export_excel(self, df, file_name=\"data_export\"):\n \"\"\"\u5bfc\u51faExcel\"\"\"\n save_path = os.path.join(REPORT_DIR, f\"{file_name}.xlsx\")\n df.to_excel(save_path, index=False)\n logger.info(f\"Excel\u6587\u4ef6\u5df2\u5bfc\u51fa\uff1a{save_path}\")\n return save_path\n\nreport_engine = ReportGenerator()\n<\/code><\/pre>\n\n\n\n10. FastAPI \u63a5\u53e3 api\/route.py\uff08\u5bf9\u5916\u63a5\u53e3 + OpenClaw \u5bf9\u63a5\uff09<\/h3>\n\n\n\n
from fastapi import FastAPI, Query\nfrom data_source.data_loader import data_loader\nfrom core.data_clean import data_cleaner\nfrom factor.finance_factor import factor_calc\nfrom statistics.analysis import finance_analyst\nfrom risk_warn.warning import warn_engine\nfrom report.report_generator import report_engine\n\napp = FastAPI(title=\"Finance \u91d1\u878d\u6570\u636e\u5206\u6790\u5e73\u53f0\")\n\n@app.get(\"\/daily_analysis\")\ndef daily_analysis(\n code: str = Query(..., description=\"\u6807\u7684\u4ee3\u7801\"),\n start: str = Query(\"2024-01-01\"),\n end: str = Query(\"2026-06-01\")\n):\n \"\"\"\u5355\u6807\u7684\u5168\u6d41\u7a0b\u5206\u6790\uff1a\u884c\u60c5+\u6e05\u6d17+\u56e0\u5b50+\u7edf\u8ba1+\u544a\u8b66\"\"\"\n # 1. \u52a0\u8f7d\u6570\u636e\n df = data_loader.get_daily_bar(code, start, end)\n if df is None:\n return {\"status\": \"fail\", \"msg\": \"\u6570\u636e\u52a0\u8f7d\u5931\u8d25\"}\n\n # 2. \u6570\u636e\u6e05\u6d17\n df = data_cleaner.standardize_column(df)\n df = data_cleaner.fill_missing(df)\n df = data_cleaner.filter_abnormal_price(df)\n\n # 3. \u8ba1\u7b97\u56e0\u5b50\n df = factor_calc.calc_all_factors(df)\n\n # 4. \u7edf\u8ba1\u5206\u6790\n stats = finance_analyst.basic_stats(df)\n\n # 5. \u98ce\u9669\u544a\u8b66\u68c0\u6d4b\n price_warn, price_msg = warn_engine.check_price_warn(df)\n\n return {\n \"status\": \"success\",\n \"code\": code,\n \"stats\": stats,\n \"price_warning\": price_warn,\n \"warn_msg\": price_msg,\n \"data_rows\": len(df)\n }\n\n@app.get(\"\/corr\")\ndef correlation(\n code1: str = Query(...),\n code2: str = Query(...),\n start: str = Query(\"2024-01-01\"),\n end: str = Query(\"2026-06-01\")\n):\n \"\"\"\u4e24\u4e2a\u6807\u7684\u76f8\u5173\u6027\u5206\u6790\"\"\"\n df1 = data_loader.get_daily_bar(code1, start, end)\n df2 = data_loader.get_daily_bar(code2, start, end)\n if df1 is None or df2 is None:\n return {\"status\": \"fail\", \"msg\": \"\u6570\u636e\u7f3a\u5931\"}\n res = finance_analyst.corr_analysis(df1, df2)\n return {\"status\": \"success\", \"corr_result\": res}\n\n@app.get(\"\/export_report\")\ndef export_report(\n code: str = Query(...),\n start: str = Query(\"2024-01-01\"),\n end: str = Query(\"2026-06-01\")\n):\n \"\"\"\u751f\u6210\u5e76\u5bfc\u51fa\u5206\u6790\u62a5\u544a\"\"\"\n df = data_loader.get_daily_bar(code, start, end)\n if df is None:\n return {\"status\": \"fail\", \"msg\": \"\u6570\u636e\u52a0\u8f7d\u5931\u8d25\"}\n stats = finance_analyst.basic_stats(df)\n html_path = report_engine.generate_html_report(df, stats, f\"{code}_\u5206\u6790\u62a5\u544a\")\n excel_path = report_engine.export_excel(df, f\"{code}_\u539f\u59cb\u6570\u636e\")\n return {\n \"status\": \"success\",\n \"html_report\": html_path,\n \"excel_data\": excel_path\n }\n\n@app.get(\"\/health\")\ndef health():\n return {\"service\": \"Finance \u91d1\u878d\u6570\u636e\u5206\u6790\u5e73\u53f0\", \"status\": \"running\"}\n<\/code><\/pre>\n\n\n\n11. \u9879\u76ee\u5165\u53e3 main.py<\/h3>\n\n\n\n
import uvicorn\nfrom api.route import app\nfrom config.settings import API_HOST, API_PORT\nfrom utils.logger import get_logger\n\nlogger = get_logger()\n\nif __name__ == \"__main__\":\n logger.info(\"Finance \u91d1\u878d\u6570\u636e\u5206\u6790\u5e73\u53f0 \u542f\u52a8\u6210\u529f\")\n uvicorn.run(app, host=API_HOST, port=API_PORT)\n<\/code><\/pre>\n\n\n\n\u516d\u3001\u7279\u8272\u6539\u9020\u843d\u5730\u8bf4\u660e<\/h3>\n\n\n\n
\n
Memory Tiering<\/code> \u533a\u5206\u51b7\u70ed\u6570\u636e\uff0c\u5b9e\u65f6\u884c\u60c5\u3001\u5f53\u65e5\u56e0\u5b50\u5e38\u9a7b\u5185\u5b58\uff0c\u5386\u53f2\u5e74\u5ea6\u6570\u636e\u538b\u7f29\u4e0b\u6c89\uff0c\u63d0\u5347\u5927\u6570\u636e\u91cf\u8fd0\u7b97\u6548\u7387\u3002<\/li>\n\n\n\nBacktest Expert<\/code> \u505a\u7b56\u7565\u56de\u6d4b\uff1b\u5f02\u52a8\u6807\u7684\u540c\u6b65\u81f3 Stock Monitor<\/code> \u76ef\u76d8\uff1b\u5206\u6790\u62a5\u544a\u5bf9\u63a5\u89c6\u9891\u751f\u6210\u3001\u793e\u5a92\u5206\u53d1\u5de5\u5177\u3002<\/li>\n\n\n\n\u4e03\u3001\u4f7f\u7528\u793a\u4f8b<\/h3>\n\n\n\n
1. \u542f\u52a8\u670d\u52a1<\/h3>\n\n\n\n
python main.py\n<\/code><\/pre>\n\n\n\n2. \u5355\u6807\u7684\u7efc\u5408\u5206\u6790<\/h3>\n\n\n\n
http://127.0.0.1:8100\/daily_analysis?code=600519.SH&start=2024-01-01&end=2026-06-01\n<\/code><\/pre>\n\n\n\n3. \u6807\u7684\u76f8\u5173\u6027\u5206\u6790<\/h3>\n\n\n\n
http://127.0.0.1:8100\/corr?code1=600519.SH&code2=000858.SZ&start=2024-01-01&end=2026-06-01\n<\/code><\/pre>\n\n\n\n4. \u5bfc\u51fa\u5206\u6790\u62a5\u544a<\/h3>\n\n\n\n
http://127.0.0.1:8100\/export_report?code=600519.SH&start=2024-01-01&end=2026-06-01\n<\/code><\/pre>\n\n\n\n\u516b\u3001\u4e8c\u6b21\u5f00\u53d1\u62d3\u5c55\u65b9\u5411<\/h3>\n\n\n\n
\n