{"id":1569,"date":"2024-10-30T07:31:00","date_gmt":"2024-10-29T23:31:00","guid":{"rendered":"https:\/\/blog.laoyulaoyu.top\/?p=1569"},"modified":"2024-10-05T17:32:29","modified_gmt":"2024-10-05T09:32:29","slug":"%e6%89%8b%e6%8a%8a%e6%89%8b%e6%95%99%e4%bd%a0-ai-%e9%a1%be%e6%8a%95%ef%bc%9a-python-%e5%8a%a9%e5%8a%9b%ef%bc%8c%e8%bd%bb%e6%9d%be%e8%bf%88%e5%85%a5%e7%ae%97%e6%b3%95%e4%ba%a4%e6%98%93%e5%91%98","status":"publish","type":"post","link":"https:\/\/laoyulaoyu.com\/index.php\/2024\/10\/30\/%e6%89%8b%e6%8a%8a%e6%89%8b%e6%95%99%e4%bd%a0-ai-%e9%a1%be%e6%8a%95%ef%bc%9a-python-%e5%8a%a9%e5%8a%9b%ef%bc%8c%e8%bd%bb%e6%9d%be%e8%bf%88%e5%85%a5%e7%ae%97%e6%b3%95%e4%ba%a4%e6%98%93%e5%91%98\/","title":{"rendered":"\u624b\u628a\u624b\u6559\u4f60 AI \u987e\u6295\uff1a Python \u52a9\u529b\uff0c\u8f7b\u677e\u8fc8\u5165\u7b97\u6cd5\u4ea4\u6613\u5458\u4e4b\u95e8"},"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>\u672c\u6587\u4ecb\u7ecd\u4e86\u5982\u4f55\u4f7f\u7528 Python \u548c\u4e13\u4e1a\u6570\u636e\u6e90\uff08\u5982 Alpha Vantage\uff09\u6765\u83b7\u53d6\u9ad8\u8d28\u91cf\u6570\u636e\uff0c\u4ee5\u53ca\u5982\u4f55\u6784\u5efa\u548c\u5206\u6790\u91cf\u5316\u4ea4\u6613\u7b56\u7565\uff0c\u7279\u522b\u662f\u5f00\u76d8\u533a\u95f4\u7a81\u7834\uff08ORB\uff09\u7b56\u7565\uff0c\u5e76\u901a\u8fc7\u5b9e\u9645\u4ee3\u7801\u793a\u4f8b\u5c55\u793a\u4e86\u5982\u4f55\u8ba1\u7b97\u4ef7\u683c\u7f3a\u53e3\u3001\u6210\u4ea4\u91cf\u3001\u79fb\u52a8\u5e73\u5747\u7ebf\u7b49\u5173\u952e\u6307\u6807\uff0c\u4ee5\u5e2e\u52a9\u5165\u95e8\u6570\u636e\u4ea4\u6613\u8005\u505a\u51fa\u66f4\u7b80\u5355\u7684\u4ea4\u6613\u51b3\u7b56<\/mark>\u3002<\/pre>\n<\/blockquote>\n\n\n\n
\u5982\u679c\u60a8\u7740\u624b\u51c6\u5907\u5f00\u59cb\u4f7f\u7528 Python \u7f16\u5199\u80a1\u7968\u4ea4\u6613\u7b97\u6cd5\uff0c\u90a3\u4e48\u60a8\u5c06\u9762\u4e34\u7684\u6700\u7d27\u8feb\u95ee\u9898\u662f\uff1a\u6211\u9700\u8981\u4ec0\u4e48\u6837\u7684\u6570\u636e\uff0c\u6211\u5728\u54ea\u91cc\u53ef\u4ee5\u627e\u5230\u8fd9\u4e9b\u6570\u636e\u3002\u6211\u82b1\u4e86\u5f88\u591a\u65f6\u95f4\u6765\u7814\u7a76\u8fd9\u4e2a\u95ee\u9898\uff0c\u5f88\u4e50\u610f\u4e3a\u60a8\u63d0\u4f9b\u5e2e\u52a9\u3002<\/p>\n\n\n\n
\u5728\u672c\u6587\u4e2d\uff0c\u6211\u4eec\u5c06\u6784\u5efa\u4e00\u4e2a\u8d85\u7ea7\u5168\u9762\u7684\u6570\u636e\u6846\u67b6\uff0c\u5176\u4e2d\u5305\u542b\u521b\u5efa\u5236\u80dc\u7b97\u6cd5\u6240\u9700\u7684\u4e00\u5207\u3002\u4f46\u66f4\u91cd\u8981\u7684\u662f\uff0c\u6211\u5728\u8fd9\u91cc\u5206\u4eab\u7684\u4ee3\u7801\u8bbe\u8ba1\u5f97\u975e\u5e38\u5bb9\u6613\u9002\u5e94\u4e0d\u540c\u7684\u65f6\u95f4\u6846\u67b6\uff0c\u5e0c\u671b\u4f60\u5728\u9700\u8981\u6293\u53d6\u81ea\u5df1\u7684\u6570\u636e\u65f6\u53ef\u4ee5\u91cd\u590d\u4f7f\u7528\u3002<\/p>\n\n\n\n
\u5e02\u9762\u4e0a\u6709\u5f88\u591a\u80a1\u7968\u6570\u636e API\uff0c\u4f46\u5bf9\u4e8e\u5b9a\u91cf\u5206\u6790\uff0c\u6211\u63a8\u8350 https:\/\/www.alphavantage.co\/<\/a>\uff0c\u56e0\u4e3a\u5b83\u5e26\u6765\u4e86\u5f88\u591a\u5df2\u7ecf\u8ba1\u7b97\u597d\u7684\u6307\u6807\uff0c\u800c\u4e14\u76d8\u4e2d\u7c92\u5ea6\u53ef\u8fbe 1 \u5206\u949f\uff0c\u8fd9\u662f\u5b9a\u91cf\u5206\u6790\u4eba\u5458\u7684\u5fc5\u5907\u5de5\u5177\u3002<\/p>\n\n\n\n
\u5177\u4f53\u5185\u5bb9\u53ef\u4ee5\u770b\u6211\u8fd9\u7bc7\u6587\u7ae0\uff1a\u300a\u624b\u628a\u624b\u6559\u4f60ai\u987e\u6295\uff1a\u53d1\u73b0\u4e00\u4e2a\u9ad8\u8d28\u91cf\u7684\u91d1\u878d\u6570\u636e\u6e90<\/a>\u300b<\/p>\n\n\n\n
\u4e00\u3001\u83b7\u53d6\u6570\u636e<\/strong><\/h3>\n\n\n\n
\u4e0b\u9762\u7684\u4ee3\u7801\u5c06\u81ea\u52a8\u6293\u53d6\u60a8\u9009\u62e9\u7684\u80a1\u7968\u6570\u636e\uff0c\u5e76\u5c06\u5b83\u4eec\u8fde\u63a5\u5230\u4e00\u4e2a\u6570\u636e\u6846\u4e2d\uff1a<\/p>\n\n\n\n
!pip install alpha_vantage\nfrom datetime import datetime\nfrom datetime import timedelta\n\nfrom alpha_vantage.timeseries import TimeSeries\nimport requests\nimport pandas as pd\nimport numpy as np\nimport datetime\nfrom re import M\nimport requests\napi_key = 'MY_ALPHA_VANTAGE_KEY'\n\ndef get_last_n_years_months(n):\n    current_date = datetime.now()\n    start_date = current_date - timedelta(days=n*365)  # Approximate start date n years ago\n    months = []\n\n    while start_date <= current_date:\n        month = start_date.strftime('%Y-%m')\n        months.append(month)\n        start_date += timedelta(days=32)  # Move to the next month\n        start_date = start_date.replace(day=1)  # Set to the first day of the next month\n\n    return months\n\ndef fetch_stock_data(ticker, n_years, api_key):\n    period = get_last_n_years_months(n_years)\n    final_df = pd.DataFrame()\n\n    for i in period:\n        print(f'Getting data for stock {ticker} and period {i}')\n        url = f'https:\/\/www.alphavantage.co\/query?function=TIME_SERIES_INTRADAY&symbol={ticker}&interval=5min&month={i}&outputsize=full&apikey={api_key}'\n        r = requests.get(url)\n\n        if r.status_code == 200:\n            data = r.json()\n            if 'Time Series (5min)' in data:\n                time_series = data['Time Series (5min)']\n                df = pd.DataFrame.from_dict(time_series, orient='index')\n                df.index = pd.to_datetime(df.index)\n                df.columns = ['open', 'high', 'low', 'close', 'volume']\n                df = df.apply(pd.to_numeric)\n                df['ticker'] = ticker  # Add ticker column to identify the stock\n                final_df = pd.concat([final_df, df])\n            else:\n                print(f\"No 'Time Series (5min)' data found for {i}\")\n        else:\n            print(f\"Failed to fetch data for {i}, status code: {r.status_code}\")\n\n    # Filter out data when market is closed\n    final_df = final_df.between_time('09:30', '15:55')\n\n    return final_df\n\n# Function to fetch and concatenate data for multiple tickers\ndef fetch_and_concatenate_data(tickers, n_years, api_key):\n    combined_df = pd.DataFrame()\n    for ticker in tickers:\n        df = fetch_stock_data(ticker, n_years, api_key)\n        combined_df = pd.concat([combined_df, df])\n    return combined_df<\/code><\/code><\/pre>\n\n\n\n
\u9996\u5148\u6211\u4eec\u521b\u5efa\u4e86\u4e00\u4e2a\u51fd\u6570 “get_last_n_years_months”\uff0c\u8be5\u51fd\u6570\u83b7\u53d6\u8fc7\u53bb\u82e5\u5e72\u5e74\u7684\u6570\u636e\u5e76\u521b\u5efa\u6708\u65e5\u671f\uff0c\u4ee5\u4fbf\u5728 API \u8c03\u7528\u4e2d\u4f7f\u7528\u3002\u901a\u5e38\uff0c\u5728\u8fdb\u884c\u5386\u53f2\u5206\u6790\u65f6\uff0c\u6700\u597d\u4f7f\u7528\u8db3\u591f\u591a\u7684\u5e74\u6570\uff0c\u4ee5\u5305\u62ec\u51e0\u4e2a\u718a\u5e02\u548c\u725b\u5e02\uff0c\u6211\u559c\u6b22\u81f3\u5c11\u9009\u62e9 5 \u5e74\u8de8\u5ea6\u3002<\/p>\n\n\n\n
\u4e4b\u540e\uff0c\u6211\u4eec\u4f7f\u7528 “fetch_stock_data “\u51fd\u6570\u8c03\u7528\u5e94\u7528\u7a0b\u5e8f\u63a5\u53e3\uff0c\u5728\u672c\u4f8b\u4e2d\uff0c\u6211\u4eec\u5c06\u4f7f\u7528 5 \u5206\u949f\u95f4\u9694\u7684\u6570\u636e\uff0c\u4f46\u60a8\u4e5f\u53ef\u4ee5\u6839\u636e\u60f3\u8981\u521b\u5efa\u7684\u7b97\u6cd5\u7c7b\u578b\u9009\u62e9\u4efb\u4f55\u4e0d\u540c\u7684\u95f4\u9694\u3002<\/p>\n\n\n\n
\u6700\u540e\uff0c”fetch_and_concatenate_data “\u51fd\u6570\u5c06\u6570\u636e\u8fde\u63a5\u5230\u4e00\u4e2a\u6570\u636e\u5e27\u4e2d\u3002<\/p>\n\n\n\n
\u4e8c\u3001\u8ba1\u7b97\u6bcf\u65e5\u95f4\u9699\u548c\u4ea4\u6613\u91cf<\/strong><\/h3>\n\n\n\n
\u5728\u591a\u6570\u5206\u6790\u91cc\uff0c\u6709\u4e2a\u7279\u522b\u91cd\u8981\u7684\u73af\u8282\u5c31\u662f\u7b97\u51fa\u76d8\u524d\u8df3\u7a7a\u7f3a\u53e3\uff0c\u4e5f\u5c31\u662f\u80a1\u7968\u4ef7\u683c\u548c\u6210\u4ea4\u91cf\u8ddf\u4e0a\u4e2a\u4ea4\u6613\u65e5\u6536\u76d8\u4ef7\u4e4b\u95f4\u7684\u5dee\u503c\u3002\u5c24\u5176\u662f\u5728\u5404\u79cd\u65e5\u5185\u4ea4\u6613\u7b97\u6cd5\u91cc\uff0c\u4f60\u5f97\u7559\u610f\u7f3a\u53e3\u7684\u53d8\u5316\u3002<\/p>\n\n\n\n
\u8bf4\u5230\u5e95\uff0c\u6210\u529f\u7684\u65e5\u95f4\u4ea4\u6613\u8005\u5e76\u4e0d\u662f\u5f04\u51fa\u4e2a\u6bcf\u5929\u90fd\u80fd\u5b8c\u7f8e\u4ea4\u6613\u80a1\u7968\u7684\u7b97\u6cd5\u3002\u4ed6\u4eec\u5f00\u53d1\u7684\u662f\u80fd\u5728\u4efb\u4f55\u7279\u5b9a\u65f6\u5019\u627e\u5230\u5b8c\u7f8e\u80a1\u7968\u7684\u7b97\u6cd5\u3002\u4e3a\u5565\u5462\uff0c\u56e0\u4e3a\u5f88\u591a\u65f6\u5019\uff0c\u67d0\u53ea\u80a1\u7968\u6216\u67d0\u4e2a\u6307\u6570\u51e0\u4e4e\u90fd\u6ca1\u4ec0\u4e48\u53d8\u5316\uff0c\u4e0d\u7ba1\u4f60\u7684\u7b97\u6cd5\u591a\u725b\uff0c\u5c31\u7b97\u4f60\u5224\u65ad\u5bf9\u4e86\u53d8\u52a8\u65b9\u5411\uff0c\u76c8\u5229\u4e5f\u4f1a\u5f88\u5c11\u3002<\/em><\/p>\n\n\n\n
\u8981\u5f00\u53d1\u7b97\u6cd5\u6765\u627e\u5230\u6bcf\u5929\u4ea4\u6613\u7684\u5b8c\u7f8e\u80a1\u7968\uff0c\u5c31\u9700\u8981\u67e5\u770b\u76d8\u524d\u7684\u4ef7\u683c\u548c\u6210\u4ea4\u91cf\u8d70\u52bf\uff0c\u627e\u5230\u53ef\u80fd\u5728\u5f53\u524d\u4ea4\u6613\u65f6\u6bb5\u5927\u5e45\u6ce2\u52a8\u3001\u5e26\u6765\u5de8\u5927\u83b7\u5229\u673a\u4f1a\u7684\u80a1\u7968\u3002<\/p>\n\n\n\n
\u6211\u4eec\u901a\u8fc7 Python \u51fd\u6570\u83b7\u53d6\u66f4\u591a\u4fe1\u606f\uff1a<\/p>\n\n\n\n
# Calculate the gap and previous_volume\ndef calculate_gap_and_previous_volume(df):\n    gap_list = []\n    previous_volume_list = []\n    for ticker in df['ticker'].unique():\n        ticker_df = df[df['ticker'] == ticker]\n        ticker_df = ticker_df.sort_index()\n\n        daily_volume = ticker_df['volume'].resample('D').sum()\n        daily_volume = daily_volume[daily_volume > 0]  # Exclude days with zero volume\n        avg_daily_volume = daily_volume.mean()\n\n        for date in ticker_df.index.normalize().unique():\n            day_data = ticker_df[ticker_df.index.normalize() == date]\n            if not day_data.empty:\n                open_price = day_data['open'].iloc[0]\n                previous_close = ticker_df[ticker_df.index < date]['close'].iloc[-1] if not ticker_df[ticker_df.index < date].empty else float('nan')\n                gap = (open_price - previous_close) \/ previous_close * 100 if not pd.isna(previous_close) else float('nan')\n                gap_list.append({'ticker': ticker, 'date': date, 'gap': gap})\n\n                # Find the last trading day with non-zero volume\n                prev_date = date - timedelta(days=1)\n                while prev_date not in daily_volume.index or daily_volume[prev_date] == 0:\n                    prev_date -= timedelta(days=1)\n                    if prev_date < daily_volume.index.min():\n                        prev_date = pd.NaT\n                        break\n                previous_day_volume = daily_volume[prev_date] if prev_date in daily_volume.index and not pd.isna(prev_date) else float('nan')\n                previous_volume = previous_day_volume \/ avg_daily_volume if not pd.isna(previous_day_volume) else float('nan')\n                previous_volume_list.append({'ticker': ticker, 'date': date, 'previous_volume': previous_volume})\n\n    gap_df = pd.DataFrame(gap_list)\n    previous_volume_df = pd.DataFrame(previous_volume_list)\n    df = df.reset_index().merge(gap_df, how='left', left_on=['ticker', df.index.normalize()], right_on=['ticker', 'date']).set_index('index')\n    df = df.reset_index().merge(previous_volume_df, how='left', left_on=['ticker', df.index.normalize()], right_on=['ticker', 'date']).set_index('index')\n\n    # Clean up merged columns\n    df = df.drop(columns=['date_x', 'date_y'])\n\n    return df<\/code><\/code><\/pre>\n\n\n\n
\u8ba9\u6211\u4eec\u62ff\u4e00\u4e9b\u70ed\u95e8\u80a1\u7968\uff08\u7279\u65af\u62c9\u3001\u82f1\u4f1f\u8fbe\u548c Meta \u80a1\u7968\uff09\u6765\u505a\u4f8b\u5b50\uff1a<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\u4f7f\u7528\u8fd9\u4e9b\u51fd\u6570\u540e\uff0c\u6211\u4eec\u5c31\u5f97\u5230\u4e86\u8fd9\u4e2a\u6570\u636e\u5e27\uff08\u5305\u542b\u524d\u4e00\u5929\u4ef7\u683c\u5dee\u8ddd\u548c\u76f8\u5bf9\u65e5\u6210\u4ea4\u91cf\u7684\u6570\u636e\uff09\uff1a<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n

\u4e09\u3001\u8ba1\u7b97\u57fa\u672c\u6307\u6807<\/strong><\/h3>\n\n\n\n
\u8ba9\u6211\u4eec\u7528 Python \u8ba1\u7b97\u4e00\u4e9b\u5e38\u7528\u4e8e\u5b9a\u91cf\u5206\u6790\u7684\u6700\u91cd\u8981\u7684\u6280\u672f\u6307\u6807\u3002\u5728\u672c\u793a\u4f8b\u4e2d\uff0c\u6211\u4eec\u5c06\u8ba1\u7b97\u5f00\u76d8\u7a81\u7834\u7b56\u7565\u7684\u6307\u6807\uff0c\u8be5\u7b56\u7565\u662f\u6700\u8457\u540d\u3001\u6700\u8d5a\u94b1\u7684\u65e5\u5185\u4ea4\u6613\u6280\u5de7\u4e4b\u4e00\u3002\u8be5\u7b56\u7565\u5305\u62ec\u5728\u5f00\u76d8\u533a\u95f4\u7684\u6da8\u8dcc\u5e45\u9650\u5236\u65f6\u4e70\u5165\u80a1\u7968\uff0c\u53ef\u7528\u4e8e 2\u30015\u300115\u300130 \u6216 60 \u5206\u949f\u7684\u533a\u95f4\u3002<\/p>\n\n\n\n
\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u4ee3\u7801\u8ba1\u7b97\u79fb\u52a8\u5e73\u5747\u7ebf\u548c\u6210\u4ea4\u91cf\u52a0\u6743\u79fb\u52a8\u5e73\u5747\u7ebf\uff0c\u4ee5\u53ca\u5f00\u76d8\u533a\u95f4\u7684\u6307\u6807\uff0c\u5982\u6210\u4ea4\u91cf\u3001\u5bbd\u5ea6\u3001\u6ce2\u52a8\u7387\u7b49\uff0c\u8fd9\u4e9b\u6307\u6807\u5bf9\u4e8e\u4e86\u89e3\u5f00\u76d8\u540e\u7b2c\u4e00\u5206\u949f\u5185\u80a1\u7968\u7684\u8d70\u52bf\u975e\u5e38\u91cd\u8981\u3002<\/p>\n\n\n\n
# Prepare the DataFrame with necessary calculations\ndef calculate_moving_averages(df):\n    df['MA_5'] = df['close'].rolling(window=5).mean()\n    df['MA_10'] = df['close'].rolling(window=10).mean()\n    df['MA_20'] = df['close'].rolling(window=20).mean()\n\n    # Adding the distance from moving averages as percentages\n    df['distance_MA_5_pct'] = (df['close'] - df['MA_5']) \/ df['MA_5'] * 100\n    df['distance_MA_10_pct'] = (df['close'] - df['MA_10']) \/ df['MA_10'] * 100\n    df['distance_MA_20_pct'] = (df['close'] - df['MA_20']) \/ df['MA_20'] * 100\n\n    return df\n\ndef calculate_vwap(df):\n    df['VWAP'] = (df['volume'] * (df['high'] + df['low'] + df['close']) \/ 3).cumsum() \/ df['volume'].cumsum()\n    return df\n\ndef prepare_dataframe(df):\n    df.index = pd.to_datetime(df.index)\n    df = calculate_moving_averages(df)\n    df = calculate_vwap(df)\n    return df\n\n# Function to calculate indicators for each stock\ndef prepare_indicators(ticker, df, opening_range_duration):\n    opening_range_df_list = []\n    for day in df.index.normalize().unique():\n        day_data = df[df.index.normalize() == day]\n\n        # Ensure data is sorted by timestamp\n        day_data = day_data.sort_index()\n\n        if len(day_data) < opening_range_duration:\n            continue\n\n        # Select the correct opening range\n        opening_range = day_data.iloc[:opening_range_duration]\n        opening_range_high = opening_range['high'].max()\n        opening_range_low = opening_range['low'].min()\n        opening_range_close = opening_range['close'].iloc[-1]\n        opening_price = opening_range['open'].iloc[0]\n        opening_range_width = ((opening_range_high - opening_range_low) \/ opening_price) * 100\n        opening_range_volatility = opening_range['close'].std() if opening_range_duration > 1 else 0\n        opening_range_volume = opening_range['volume'].sum()\n        opening_range_avg_volume = opening_range_volume \/ opening_range_duration\n\n        post_opening_range = day_data.iloc[opening_range_duration:]\n\n        breakout_idx = None\n        breakout_type = None\n        for idx in post_opening_range.index:\n            if post_opening_range.loc[idx]['low'] > opening_range_high:\n                breakout_idx = idx\n                breakout_type = 'bullish'\n                break\n            elif post_opening_range.loc[idx]['high'] < opening_range_low:\n                breakout_idx = idx\n                breakout_type = 'bearish'\n                break\n\n        if breakout_idx is None:\n            continue\n\n        breakout_data = day_data.loc[breakout_idx]\n\n        next_1_hour = day_data.loc[breakout_idx:].iloc[1:13]\n        if breakout_type == 'bullish':\n            max_price_next_1_hour = next_1_hour['high'].max() if not next_1_hour.empty else np.nan\n            target = (max_price_next_1_hour - breakout_data['close']) \/ breakout_data['close'] * 100 if not np.isnan(max_price_next_1_hour) else np.nan\n        else:\n            min_price_next_1_hour = next_1_hour['low'].min() if not next_1_hour.empty else np.nan\n            target = (breakout_data['close'] - min_price_next_1_hour) \/ breakout_data['close'] * 100 if not np.isnan(min_price_next_1_hour) else np.nan\n\n        # Calculate percentage change of breakout close price from opening range close price\n        price_change_during_breakout = (breakout_data['close'] - opening_range_close) \/ opening_range_close * 100\n\n        # Calculate direction consistency in the opening range\n        if breakout_type == 'bullish':\n            direction_consistency = (opening_range['close'] > opening_range['open']).sum() \/ len(opening_range)\n        else:\n            direction_consistency = (opening_range['close'] < opening_range['open']).sum() \/ len(opening_range)\n\n        # Calculate volume relative to the average volume during the opening range\n        volume_relative_to_avg = (opening_range_volume - opening_range_avg_volume) \/ opening_range_avg_volume * 100\n\n        # Calculate VWAP as a percentage relative to the close price\n        vwap_end_pct = (breakout_data['VWAP'] - breakout_data['close']) \/ breakout_data['close'] * 100\n        vwap_at_breakout_pct = (breakout_data['VWAP'] - breakout_data['close']) \/ breakout_data['close'] * 100\n\n        row_data = {\n            'ticker': ticker,\n            'date': day,\n            'opening_range_width': opening_range_width,\n            'opening_range_volatility': opening_range_volatility,\n            'direction_consistency': direction_consistency,\n            'gap': df.loc[day_data.index[0], 'gap'],\n            'previous_volume': df.loc[day_data.index[0], 'previous_volume'],\n            'price_change_during_breakout': price_change_during_breakout,\n            'volume_relative_to_avg': volume_relative_to_avg,\n            'distance_MA_5_pct': breakout_data['distance_MA_5_pct'],\n            'distance_MA_10_pct': breakout_data['distance_MA_10_pct'],\n            'distance_MA_20_pct': breakout_data['distance_MA_20_pct'],\n            'VWAP_End_pct': vwap_end_pct,\n            'VWAP_at_Breakout_pct': vwap_at_breakout_pct,\n            'breakout_type': breakout_type,\n            'target': target\n        }\n\n        opening_range_df_list.append(row_data)\n\n    return opening_range_df_list\n\n\n# Apply the prepare_indicators function to each stock\ndef analyze_combined_data(df, opening_range_duration):\n    opening_range_df_list = []\n    for ticker in df['ticker'].unique():\n        ticker_df = df[df['ticker'] == ticker]\n        opening_range_df_list.extend(prepare_indicators(ticker, ticker_df, opening_range_duration))\n    return pd.DataFrame(opening_range_df_list)<\/code><\/code><\/pre>\n\n\n\n
\u8ba9\u6211\u4eec\u770b\u770b\u5b83\u5982\u4f55\u4f7f\u7528 15 \u5206\u949f\u5f00\u4ed3\u533a\u95f4\u7b56\u7565\u7684\u6570\u636e\uff08\u5373 5 \u5206\u949f\u6570\u636e\u7684 3 \u4e2a\u533a\u95f4\uff09\uff1a<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\u5982\u60a8\u6240\u89c1\uff0c\u6211\u4eec\u5728\u4e0a\u9762\u6784\u5efa\u4e86\u4e00\u4e2a\u975e\u5e38\u5168\u9762\u7684\u6570\u636e\u6846\u67b6\uff0c\u4f7f\u7528\u4e86\u79fb\u52a8\u5e73\u5747\u7ebf\u7b49\u5e38\u7528\u6307\u6807\uff0c\u4ee5\u53ca\u5176\u4ed6\u5bf9\u6211\u4eec\u6709\u610f\u4e49\u7684\u6307\u6807\u3002\u5728\u7814\u7a76\u4e2d\uff0c\u4f60\u53ea\u9700\u95ee\u95ee\u81ea\u5df1\uff0c\u4f60\u60f3\u6d4b\u8bd5\u7684\u5047\u8bbe\u662f\u4ec0\u4e48\uff1f\u5728\u8fd9\u79cd\u60c5\u51b5\u4e0b\uff0c\u6211\u4eec\u8981\u6d4b\u8bd5\u7684\u4fe1\u5ff5\u5c31\u662f ORB \u7b56\u7565\u4e2d\u6240\u5305\u542b\u7684\u4fe1\u5ff5\u3002<\/p>\n\n\n\n
\u8bf7\u6ce8\u610f\uff0c\u4e3a\u8fbe\u5230\u76ee\u7684\uff0c\u6211\u521b\u5efa\u4e86\u4e00\u4e2a\u76ee\u6807\uff0c\u7528\u6765\u8861\u91cf\u6211\u4eec\u7528\u8fd9\u884c\u4ee3\u7801\u4e70\u5165\u770b\u6da8\u7a81\u7834\u6216\u5356\u51fa\u770b\u8dcc\u7a81\u7834\u540e\uff0c\u5728\u63a5\u4e0b\u6765\u7684 60 \u5206\u949f\u5185\u53ef\u4ee5\u83b7\u5f97\u7684\u6700\u5927\u5229\u6da6\uff1a<\/p>\n\n\n\n
target = (breakout_data['close'] - min_price_next_1_hour) \/ breakout_data['close'] * 100 if not np.isnan(min_price_next_1_hour) else np.nan<\/code><\/code><\/pre>\n\n\n\n
\u60a8\u53ef\u4ee5\u6839\u636e\u81ea\u5df1\u7684\u559c\u597d\u8fdb\u884c\u8c03\u6574\uff0c\u4f46\u901a\u5e38 ORB \u5305\u62ec\u5728\u7a81\u7834\u540e\u77ed\u65f6\u95f4\u5185\u6301\u4ed3\u3002\u5982\u679c\u4ea4\u6613\u6210\u529f\uff08\u80a1\u7968\u5411\u60a8\u671f\u671b\u7684\u65b9\u5411\u79fb\u52a8\uff09\uff0c\u5c31\u5fc5\u987b\u51b3\u5b9a\u4f55\u65f6\u79bb\u573a\u3002<\/p>\n\n\n\n
\u76f8\u53cd\uff0c\u5982\u679c\u4ea4\u6613\u4e0e\u60a8\u7684\u9884\u671f\u80cc\u9053\u800c\u9a70\uff0c\u60a8\u5c31\u5fc5\u987b\u51b3\u5b9a\u4f55\u65f6\u51fa\u5c40\uff0c\u5728\u635f\u5931\u8fc7\u5927\u4e4b\u524d\u51cf\u5c11\u635f\u5931\u3002<\/p>\n\n\n\n
\u4e00\u822c\u6765\u8bf4\uff0c\u65e5\u95f4\u4ea4\u6613\u8005\u4f1a\u9009\u62e9 1:3 \u7684\u98ce\u9669\u4e0e\u56de\u62a5\u6bd4\u4f8b\uff0c\u8fd9\u53ef\u80fd\u8868\u793a\u4ed6\u4eec\u4f1a\u5728\u4e8f\u635f 1%\u7684\u65f6\u5019\u5e73\u6389\u4ed3\u4f4d\uff0c\u5728\u76c8\u5229\u8fbe\u5230 3%\u7684\u65f6\u5019\u83b7\u5229\u4e86\u7ed3\uff0c\u6216\u8005\u4ed6\u4eec\u4f1a\u5728\u4e8f\u635f\u8d85\u8fc7 0.5%\u7684\u65f6\u5019\u5c31\u5e73\u4ed3\uff0c\u5728\u76c8\u5229\u8d85\u8fc7 1.5%\u7684\u65f6\u5019\u8d5a\u4e00\u7b14\u3002<\/em><\/p>\n\n\n\n
\u5c31\u662f\u4e3a\u4f55\u6211\u4eec\u8981\u7814\u7a76\u4f60\u6240\u80fd\u83b7\u53d6\u7684\u6700\u5927\u5229\u6da6\uff0c\u56e0\u4e3a\u8fd9\u662f\u51b3\u5b9a\u4f55\u4e3a\u5408\u7406\u79bb\u573a\u4ef7\u683c\u3001\u5bf9\u5229\u6da6\u611f\u5230\u6ee1\u8db3\u4ee5\u53ca\u4f55\u79cd\u7a0b\u5ea6\u7684\u4e8f\u635f\u662f\u6211\u4eec\u80fd\u591f\u63a5\u53d7\u6216\u65e0\u6cd5\u63a5\u53d7\u7684\u5173\u952e\u8981\u7d20\u3002<\/p>\n\n\n\n
\u901a\u8fc7\u6570\u636e\u7814\u7a76\uff0c\u6211\u4eec\u53ef\u4ee5\u786e\u5b9a\u4ec0\u4e48\u662f\u771f\u6b63\u7684\u6700\u4f73\u503c\uff0c\u5e76\u6839\u636e\u56de\u6eaf\u6d4b\u8bd5\u505a\u51fa\u660e\u667a\u7684\u51b3\u5b9a\u3002<\/p>\n\n\n\n
\u56db\u3001\u89c2\u70b9\u603b\u7ed3<\/strong><\/h3>\n\n\n\n
\u73b0\u5728\uff0c\u60a8\u4e0d\u5fc5\u56e0\u4e3a\u535a\u4e3b\u8fd9\u4e48\u8bf4\uff0c\u5c31\u575a\u4fe1\u5f00\u4ed3\u533a\u95f4\u7a81\u7834\uff08\u6216\u4efb\u4f55\u5176\u4ed6\u7b56\u7565\uff09\u662f\u6709\u6548\u7684\u3002\u6b63\u786e\u7684\u91cf\u5316\u601d\u7ef4\u5e94\u5efa\u7acb\u5728\u5bf9\u6570\u636e\u53ca\u81ea\u8eab\u7814\u7a76\u5145\u6ee1\u4fe1\u5fc3\u7684\u57fa\u77f3\u4e4b\u4e0a\u3002\u90a3\u4e48\uff0c\u5c31\u8ba9\u6211\u4eec\u4eb2\u81ea\u6765\u77a7\u77a7\u8fd9\u4e2a\u7cbe\u7f8e\u7684\u6570\u636e\u6846\u67b6\u6240\u80fd\u544a\u77e5\u6211\u4eec\u7684\u5185\u5bb9\u5427\u3002<\/p>\n\n\n\n
\u5728\u7406\u60f3\u7684\u60c5\u51b5\u4e0b\uff0c\u5982\u679c\u4e00\u53ea\u80a1\u7968\u5728\u5f00\u76d8\u524d\u5927\u5e45\u8df3\u7a7a\u4e0a\u6da8\uff0c\u6210\u4ea4\u91cf\u5f3a\u52b2\uff0c\u5e76\u4e14\u5728\u4e0a\u5348\u7ee7\u7eed\u671d\u540c\u4e00\u65b9\u5411\u4e0a\u626c\uff0c\u6ca1\u6709\u4e1d\u6beb\u72b9\u8c6b\uff08\u8fd9\u5c31\u662f\u6211\u4eec\u6d4b\u91cf\u6ce2\u52a8\u7387\u548c\u65b9\u5411\u4e00\u81f4\u6027\u7684\u539f\u56e0\uff09\uff0c\u90a3\u4e48\u8fd9\u53ea\u80a1\u7968\u5c31\u5f88\u6709\u53ef\u80fd\u5728\u5f00\u76d8\u540e\u7ee7\u7eed\u4fdd\u6301\u8fd9\u79cd\u8d70\u52bf\u3002\u73b0\u5728\uff0c\u6211\u4eec\u60f3\u77e5\u9053\u6570\u636e\u662f\u5426\u652f\u6301\u8fd9\u4e00\u7406\u8bba\uff0c\u6211\u4eec\u60f3\u77e5\u9053\u8fd9\u4e9b\u80a1\u7968\u6309\u7167\u6211\u4eec\u5047\u8bbe\u7684\u8d70\u52bf\u8fd0\u884c\u7684\u6982\u7387\u3002<\/p>\n\n\n\n
\u6211\u4eec\u9700\u8981\u63a2\u7d22\u53d8\u91cf\u548c\u76ee\u6807\u4e4b\u95f4\u7684\u5173\u7cfb\uff0c\u770b\u770b\u5b83\u4eec\u662f\u5426\u771f\u7684\u6709\u5f71\u54cd\uff0c\u66f4\u91cd\u8981\u7684\u662f\uff0c\u6211\u4eec\u4e0d\u4ec5\u8981\u9884\u6d4b\u76ee\u6807\uff0c\u8fd8\u8981\u8861\u91cf\u6211\u4eec\u7684\u9884\u6d4b\u7ed3\u679c\u6709\u591a\u597d\u3002<\/p>\n\n\n\n
    \n
  1. \u83b7\u53d6\u9ad8\u8d28\u91cf\u6570\u636e\u662f\u91cf\u5316\u4ea4\u6613\u6210\u529f\u7684\u5173\u952e<\/strong>\u3002Alpha Vantage \u63d0\u4f9b\u4e86\u4e30\u5bcc\u7684\u80a1\u7968\u6570\u636e\u548c\u5df2\u8ba1\u7b97\u7684\u6307\u6807\uff0c\u662f\u91cf\u5316\u5206\u6790\u7684\u7406\u60f3\u9009\u62e9\u3002<\/li>\n\n\n\n
  2. \u4ef7\u683c\u7f3a\u53e3\u548c\u6210\u4ea4\u91cf\u662f\u65e5\u5185\u4ea4\u6613\u51b3\u7b56\u7684\u91cd\u8981\u6307\u6807<\/strong>\u3002\u5b83\u4eec\u53ef\u4ee5\u5e2e\u52a9\u4ea4\u6613\u8005\u8bc6\u522b\u90a3\u4e9b\u5728\u5f00\u76d8\u540e\u53ef\u80fd\u51fa\u73b0\u5927\u5e45\u6ce2\u52a8\u7684\u80a1\u7968\u3002<\/li>\n\n\n\n
  3. \u5f00\u76d8\u533a\u95f4\u7a81\u7834\uff08ORB\uff09\u7b56\u7565\u662f\u4e00\u79cd\u6709\u6548\u7684\u65e5\u5185\u4ea4\u6613\u7b56\u7565<\/strong>\u3002\u901a\u8fc7\u5206\u6790\u5f00\u76d8\u533a\u95f4\u7684\u5bbd\u5ea6\u3001\u6ce2\u52a8\u6027\u3001\u6210\u4ea4\u91cf\u548c\u65b9\u5411\u4e00\u81f4\u6027\u7b49\u6307\u6807\uff0c\u53ef\u4ee5\u66f4\u597d\u5730\u9884\u6d4b\u80a1\u7968\u7684\u8d70\u52bf\u3002<\/li>\n\n\n\n
  4. \u98ce\u9669\u4e0e\u56de\u62a5\u6bd4\u7387\u662f\u4ea4\u6613\u51b3\u7b56\u7684\u91cd\u8981\u8003\u91cf<\/strong>\u3002\u65e5\u95f4\u4ea4\u6613\u8005\u901a\u5e38\u4f1a\u9009\u62e9 1:3 \u7684\u98ce\u9669\u4e0e\u56de\u62a5\u6bd4\u7387\uff0c\u5e76\u6839\u636e\u5b9e\u9645\u6570\u636e\u7814\u7a76\u6765\u51b3\u5b9a\u4f55\u65f6\u8fdb\u51fa\u5e02\u573a\u3002<\/li>\n<\/ol>\n\n\n\n
    <\/p>\n\n\n\n
    \u5ef6\u5c55\u9605\u8bfb\u63a8\u8350<\/strong><\/h3>\n\n\n\n