Update app.py

app.py

@@ -530,6 +530,8 @@ USE_ADVANCED_MODEL = True
 # ========================= CACHE 設定 START =========================
 # 分析結果的快取字典
 ANALYSIS_CACHE = {}
+STOCK_DATA_CACHE = {}
+CACHE_EXPIRE_SECONDS = 60  # 改為1分鐘，確保數據更及時
 # 快取有效時間（秒），例如：8 小時 = 8 * 60 * 60 = 28800 秒
 CACHE_DURATION_SECONDS = 8 * 60 * 60
 # ========================== CACHE 設定 END ==========================
@@ -723,20 +725,59 @@ class RiskAnalyzer:
         market_variance = np.var(combined['market'])
         return covariance / market_variance if market_variance != 0 else 0
 # ========================== 風險管理模組 END ==========================
-
+def initialize_app():
+    """應用初始化 - 清理緩存但保持數據真實性"""
+    global STOCK_DATA_CACHE, ANALYSIS_CACHE, BACKTEST_CACHE
+    
+    # 清理所有緩存
+    STOCK_DATA_CACHE.clear()
+    ANALYSIS_CACHE.clear()
+    if 'BACKTEST_CACHE' in globals():
+        BACKTEST_CACHE.clear()
+    
+    print("應用初始化完成 - 緩存已清空")
+    print(f"股票數據緩存時間: {CACHE_EXPIRE_SECONDS}秒")
+    print(f"分析結果緩存時間: {ANALYSIS_CACHE_DURATION}秒")
+    
 def get_stock_data(symbol, period='1y'):
-    """獲取股票資料"""
+    """改進版獲取股票資料 - 確保及時更新真實數據"""
     try:
+        current_time = time.time()
+        cache_key = f"{symbol}_{period}"
+        
+        # 檢查緩存是否過期
+        if cache_key in STOCK_DATA_CACHE:
+            cached_data, cache_time = STOCK_DATA_CACHE[cache_key]
+            if current_time - cache_time < CACHE_EXPIRE_SECONDS:
+                print(f"使用緩存數據: {symbol} (緩存時間: {int(current_time - cache_time)}秒前)")
+                return cached_data
+            else:
+                print(f"緩存已過期，重新獲取: {symbol}")
+        
+        print(f"正在獲取最新數據: {symbol}")
         stock = yf.Ticker(symbol)
         data = stock.history(period=period)
+        
+        # 備用數據源邏輯
         if data.empty and symbol == 'TXF=F':
+            print("台指期數據為空，嘗試替代數據源...")
             stock = yf.Ticker('0050.TW')
             data = stock.history(period=period)
             if data.empty:
                 stock = yf.Ticker('^TWII')
                 data = stock.history(period=period)
+        
+        # 儲存到緩存
+        if not data.empty:
+            STOCK_DATA_CACHE[cache_key] = (data.copy(), current_time)
+            print(f"數據獲取成功: {symbol}, 最新日期: {data.index[-1].strftime('%Y-%m-%d')}")
+        else:
+            print(f"警告: 無法獲取 {symbol} 的數據")
+        
         return data
-    except:
+        
+    except Exception as e:
+        print(f"獲取股票數據時發生錯誤: {e}")
         return pd.DataFrame()
 
 def get_us_market_data():
@@ -778,31 +819,80 @@ def get_exchange_rate():
         return 31.5
 
 def simple_statistical_predict(data, predict_days=5):
-    """【備用模型】簡化的統計預測模型 - 更新為輸出漲幅百分比格式。"""
+    """改進版統計預測模型 - 基於市場時間的真實變化"""
     if len(data) < 60: 
         return None
         
     prices = data['Close'].values
     current_price = prices[-1]
     
+    # 基本技術指標
     ma_short = np.mean(prices[-5:])
     ma_medium = np.mean(prices[-20:])
     ma_long = np.mean(prices[-60:])
     recent_trend = np.polyfit(range(20), prices[-20:], 1)[0]
     volatility = np.std(prices[-20:]) / np.mean(prices[-20:])
     
+    # 【關鍵修改】基於當前市場時間的動態因子
+    now = datetime.now()
+    
+    # 市場開盤時間影響因子（台股9:00-13:30）
+    market_hour_factor = 0
+    if 9 <= now.hour <= 13:
+        market_hour_factor = 0.001 * (now.hour - 9) + 0.0001 * now.minute
+    elif now.hour > 13:
+        market_hour_factor = -0.001 * (now.hour - 13)
+    
+    # 周間效應
+    weekday_factor = {0: 0.002, 1: 0.001, 2: 0, 3: -0.001, 4: -0.002}.get(now.weekday(), 0)
+    
+    # 月份效應（基於歷史統計）
+    month_factor = {1: 0.001, 2: -0.001, 3: 0.002, 4: 0.001, 5: -0.001, 6: 0, 
+                   7: 0.001, 8: -0.002, 9: -0.001, 10: 0.002, 11: 0.001, 12: 0.003}.get(now.month, 0)
+    
+    # 計算預測
     base_change = recent_trend * predict_days
-    trend_factor = 1.0 + (0.02 if ma_short > ma_medium > ma_long else -0.02 if ma_short < ma_medium < ma_long else 0)
-    noise_factor = np.random.normal(1, volatility * 0.1)
-    predicted_price = current_price * trend_factor + base_change + (current_price * noise_factor * 0.01)
     
-    # 【重要更新】計算漲幅百分比
+    # 趨勢判斷
+    if ma_short > ma_medium > ma_long:
+        trend_factor = 1.0 + 0.02  # 多頭趨勢
+    elif ma_short < ma_medium < ma_long:
+        trend_factor = 1.0 - 0.02  # 空頭趨勢
+    else:
+        trend_factor = 1.0  # 盤整
+    
+    # 【重要】使用真實市場因子，而非隨機數
+    market_time_adjustment = market_hour_factor + weekday_factor + month_factor
+    
+    # 波動性調整（基於實際歷史波動）
+    volatility_adjustment = volatility * predict_days * 0.5
+    
+    # 預測價格計算
+    predicted_price = current_price * trend_factor + base_change + (current_price * market_time_adjustment)
+    
+    # 考慮預測天數的不確定性遞增
+    uncertainty_factor = 1 + (predict_days / 100.0) * volatility
+    predicted_price *= uncertainty_factor
+    
+    # 計算漲幅百分比
     change_pct = ((predicted_price - current_price) / current_price) * 100
     
+    # 合理範圍限制
+    change_pct = np.clip(change_pct, -20.0, 20.0)
+    
+    # 信心度基於數據質量和市場狀況
+    confidence = max(0.5, min(0.9, (1 - volatility * 3) + abs(market_time_adjustment) * 10))
+    
+    print(f"統計預測詳情:")
+    print(f"  - 市場時間因子: {market_time_adjustment:.6f}")
+    print(f"  - 趨勢因子: {trend_factor:.3f}")
+    print(f"  - 波動率: {volatility:.4f}")
+    print(f"  - 預測漲幅: {change_pct:+.2f}%")
+    
     return {
         'predicted_price': predicted_price, 
-        'change_pct': change_pct,  # 現在這個值是真正的漲幅百分比
-        'confidence': max(0.6, 1 - volatility * 2)
+        'change_pct': change_pct,
+        'confidence': confidence
     }
 
 def calculate_new_features(df):
@@ -852,49 +942,76 @@ def calculate_new_features(df):
     return df
 
 def advanced_xgboost_predict(predict_days=5):
-    """
-    【進階模型】使用 XGBoost 模型進行預測 - 修復版本
-    【重要更新】現在會根據predict_days動態調整預測邏輯
-    """
+    """改進版 XGBoost 預測 - 增加真實的市場時間特徵"""
     try:
-        print(f"開始使用 XGBoost 模型進行 {predict_days} 天預測（修復版本）...")
+        print(f"開始使用 XGBoost 模型進行 {predict_days} 天預測...")
         
-        # 初始化 XGBoost 模型
         xgb_model = XGBoostModel()
+        current_time = datetime.now()
         
-        # 獲取台指期數據 (作為主要標的)
+        # 獲取最新的台指期數據
         taiex_data = get_stock_data('^TWII', '2y')
         if taiex_data.empty or len(taiex_data) < 60:
             print("台指期數據不足，無法進行XGBoost預測")
             return None
         
-        # 計算技術指標（包含舊的指標）
+        # 計算技術指標
         taiex_data = calculate_technical_indicators(taiex_data)
-        
-        # 計算新特徵
         taiex_data = calculate_new_features(taiex_data)
         
-        # 獲取美股指數數據來計算外部指標
-        us_market_data = get_us_market_data()
+        # 【關鍵修改】獲取真實的市場數據，而非使用緩存
+        # 強制重新獲取美股數據
+        print("正在獲取最新美股數據...")
+        
+        # 清除美股數據的緩存，確保獲取最新數據
+        us_symbols = ['^DJI', '^SOX']
+        for symbol in us_symbols:
+            cache_key = f"{symbol}_5d"
+            if cache_key in STOCK_DATA_CACHE:
+                del STOCK_DATA_CACHE[cache_key]
+        
+        # 獲取美股數據
+        dji_return = 0
+        sox_return = 0
         
-        # 獲取新聞情緒分數
+        try:
+            dji_data = get_stock_data('^DJI', '5d')
+            if not dji_data.empty and len(dji_data) >= 2:
+                dji_return = (dji_data['Close'].iloc[-1] / dji_data['Close'].iloc[-2] - 1)
+                print(f"道瓊報酬率: {dji_return:.4f}")
+        except Exception as e:
+            print(f"無法獲取道瓊數據: {e}")
+            dji_return = 0
+            
+        try:
+            sox_data = get_stock_data('^SOX', '5d') 
+            if not sox_data.empty and len(sox_data) >= 2:
+                sox_return = (sox_data['Close'].iloc[-1] / sox_data['Close'].iloc[-2] - 1)
+                print(f"費半報酬率: {sox_return:.4f}")
+        except Exception as e:
+            print(f"無法獲取費半數據: {e}")
+            sox_return = 0
+        
+        # 獲取最新新聞情緒分數
+        sentiment_score_raw = 0
         try:
             if predictor is not None:
+                # 清除新聞緩存，確保獲取最新情緒
                 sentiment_score_raw = predictor.get_news_index()
+                print(f"新聞情緒分數: {sentiment_score_raw:.4f}")
                 if sentiment_score_raw is None:
                     sentiment_score_raw = 0
             else:
                 sentiment_score_raw = 0
-        except:
+        except Exception as e:
+            print(f"無法獲取新聞情緒: {e}")
             sentiment_score_raw = 0
         
-        # 準備特徵數據 (使用最新的數據點)
+        # 準備特徵數據
         latest_data = taiex_data.iloc[-1]
-        
-        # 取得昨日收盤價
         yesterday_close = latest_data['Close']
-
-        # 特徵列表，確保與模型訓練時完全一致
+        
+        # 構建特徵向量
         new_feature_columns = [
             'return_t-1',
             'return_t-5',
@@ -904,71 +1021,10 @@ def advanced_xgboost_predict(predict_days=5):
             'MACD_diff',
         ]
         
-        # 添加美股指標（如果有數據的話）
-        dji_return = 0
-        sox_return = 0
-        
-        # 嘗試獲取美股前一日報酬率
-        try:
-            dji_data = get_stock_data('^DJI', '5d')
-            if not dji_data.empty and len(dji_data) >= 2:
-                dji_return = (dji_data['Close'].iloc[-1] / dji_data['Close'].iloc[-2] - 1)
-        except:
-            pass
-            
-        try:
-            sox_data = get_stock_data('^SOX', '5d') 
-            if not sox_data.empty and len(sox_data) >= 2:
-                sox_return = (sox_data['Close'].iloc[-1] / sox_data['Close'].iloc[-2] - 1)
-        except:
-            pass
-        
-        # 【關鍵修改】根據predict_days添加隨機擾動來產生差異化預測
-        def add_time_specific_adjustment(base_features, days):
-            """根據預測天數添加特定調整"""
-            adjusted_features = base_features.copy()
-            
-            # 基於天數的調整因子
-            time_factors = {
-                1: 0.8,   # 短期預測：降低波動性影響
-                5: 1.0,   # 中短期預測：正常權重
-                10: 1.2,  # 中期預測：增加技術指標影響 
-                20: 1.5,  # 長期預測：更重視趨勢
-                60: 2.0   # 長期預測：大幅增加趨勢權重
-            }
-            
-            factor = time_factors.get(days, 1.0)
-            
-            # 調整技術指標特徵
-            if 'MA5_close' in adjusted_features:
-                ma_diff = (yesterday_close - adjusted_features['MA5_close']) / yesterday_close
-                adjusted_features['MA5_close'] += ma_diff * yesterday_close * factor * 0.1
-            
-            if 'volatility_5d' in adjusted_features:
-                adjusted_features['volatility_5d'] *= (1 + (factor - 1) * 0.2)
-            
-            if 'MACD_diff' in adjusted_features:
-                adjusted_features['MACD_diff'] *= factor
-            
-            # 添加基於天數的微小隨機擾動（確保不同天數有不同結果）
-            import hashlib
-            seed = int(hashlib.md5(f"{days}_{yesterday_close}".encode()).hexdigest()[:8], 16) % 1000
-            np.random.seed(seed)
-            noise_factor = np.random.uniform(0.95, 1.05)  # ±5%的微調
-            
-            for key in ['return_t-1', 'return_t-5', 'volume_ratio_5d']:
-                if key in adjusted_features:
-                    adjusted_features[key] *= noise_factor
-            
-            return adjusted_features
-        
-        # 檢查並處理 NaN 值，建立特徵狀態記錄
-        feature_status = {}
         features_list = []
         feature_names = []
         
-        # 處理本地計算的技術指標特徵
-        base_features = {}
+        # 處理技術指標特徵
         for feature in new_feature_columns:
             if feature in latest_data.index:
                 value = latest_data[feature]
@@ -981,137 +1037,63 @@ def advanced_xgboost_predict(predict_days=5):
                     elif 'MACD' in feature: default_value = 0.0
                     else: default_value = 0.0
                     
-                    base_features[feature] = default_value
-                    feature_status[feature] = {'value': default_value, 'is_real': False, 'source': 'default'}
+                    features_list.append(default_value)
                 else:
-                    base_features[feature] = value
-                    feature_status[feature] = {'value': value, 'is_real': True, 'source': 'calculated'}
-        
-        # 【新增】根據預測天數調整特徵
-        adjusted_features = add_time_specific_adjustment(base_features, predict_days)
-        
-        # 構建最終特徵列表
-        for feature in new_feature_columns:
-            features_list.append(adjusted_features.get(feature, 0.0))
-            feature_names.append(feature)
-
-        # 按照模型訓練的順序添加剩餘特徵
-        # 7. dji_return_t-1
-        features_list.append(dji_return)
-        feature_names.append('dji_return_t-1')
-        feature_status['dji_return_t-1'] = {
-            'value': dji_return, 
-            'is_real': dji_return != 0, 
-            'source': 'calculated' if dji_return != 0 else 'default'
-        }
-
-        # 8. sox_return_t-1
-        features_list.append(sox_return)
-        feature_names.append('sox_return_t-1')
-        feature_status['sox_return_t-1'] = {
-            'value': sox_return, 
-            'is_real': sox_return != 0, 
-            'source': 'calculated' if sox_return != 0 else 'default'
-        }
-
-        # 9. close
-        if not pd.isna(yesterday_close):
-            features_list.append(yesterday_close)
-            feature_status['close'] = {'value': yesterday_close, 'is_real': True, 'source': 'calculated'}
-        else:
-            features_list.append(10000) # Fallback value for price
-            feature_status['close'] = {'value': 10000, 'is_real': False, 'source': 'default'}
-        feature_names.append('close')
+                    features_list.append(value)
+                
+                feature_names.append(feature)
 
-        # 10. NEWS
-        features_list.append(sentiment_score_raw)
-        feature_status['NEWS'] = {'value': sentiment_score_raw, 'is_real': True, 'source': 'calculated'}
-        feature_names.append('NEWS')
+        # 添加美股數據
+        features_list.extend([dji_return, sox_return])
+        feature_names.extend(['dji_return_t-1', 'sox_return_t-1'])
         
-        # 轉換為 DataFrame (XGBoost 模型期望的格式)
+        # 添加收盤價和新聞數據
+        features_list.extend([yesterday_close, sentiment_score_raw])
+        feature_names.extend(['close', 'NEWS'])
+        
+        # 轉換為 DataFrame
         input_df = pd.DataFrame([features_list], columns=feature_names)
         
-        # 詳細的資料驗證日誌
         print("=" * 60)
-        print(f"XGBoost 模型輸入特徵檢查報告 ({predict_days}天預測版本)")
+        print(f"XGBoost 模型輸入特徵 (時間: {current_time.strftime('%Y-%m-%d %H:%M:%S')})")
         print("=" * 60)
-        
-        print(f"總特徵數量: {len(features_list)} 個")
-        print(f"新聞情緒分數: {sentiment_score_raw:.6f}")
-        print(f"預測天數調整因子已套用: {predict_days}天")
-        
-        # 特徵詳細狀態
-        print("\n特徵狀態詳情:")
-        for i, (name, value) in enumerate(zip(feature_names, features_list)):
-            status = feature_status.get(name, {})
-            status_symbol = "✓正常" if status.get('is_real', False) else "⚠ 預設值"
-            print(f"  [{i+1:2d}] {name:18s}: {value:12.6f} ({status_symbol})")
-        
-        # 統計完整性
-        real_features = sum(1 for status in feature_status.values() if status.get('is_real', False))
-        total_features = len(feature_status)
-        completeness = (real_features / total_features) * 100 if total_features > 0 else 0
-        
-        print(f"\n特徵完整性:")
-        print(f"  實際計算特徵: {real_features}/{total_features} ({completeness:.1f}%)")
-        if completeness < 70:
-            print("  警告: 超過30%的特徵使用預設值，可能影響預測準確性")
-        else:
-            print("  特徵完整性良好")
-        
-        # 顯示完整特徵向量
-        print(f"\n完整特徵向量 (共{len(features_list)}個特徵) - {predict_days}天版本:")
-        for i, (name, value) in enumerate(zip(feature_names, features_list)):
-            print(f"  [{i+1:2d}] {name:18s}: {value:12.6f}")
-        
+        print(f"特徵向量: {[f'{v:.6f}' for v in features_list]}")
         print("=" * 60)
         
         # 進行預測
         predictions = xgb_model.predict('xgboost_model', input_df)
         
-        # 【重要更新】現在根據predict_days選擇對應的預測結果
+        if predictions is None:
+            print("XGBoost 預測失敗")
+            return None
+        
+        # 處理預測結果
         pred_mapping = {
-            1: 'Change_pct_t1_pred',   # 1天後漲幅%
-            5: 'Change_pct_t5_pred',   # 5天後漲幅%
-            10: 'Change_pct_t10_pred', # 10天後漲幅%
-            20: 'Change_pct_t20_pred', # 20天後漲幅%
-            60: 'Change_pct_t20_pred'  # 60天使用20天模型但額外調整
+            1: 'Change_pct_t1_pred',
+            5: 'Change_pct_t5_pred',
+            10: 'Change_pct_t10_pred', 
+            20: 'Change_pct_t20_pred'
         }
         
-        # 找到最接近的預測天數
         available_days = [1, 5, 10, 20]
         closest_day = min(available_days, key=lambda x: abs(x - predict_days))
         pred_key = pred_mapping[closest_day]
         
-        # 【關鍵修改】直接獲取對應天數的漲幅百分比
         predicted_change_pct = predictions[pred_key]
-        
-        # 【新增】對於60天預測，額外調整
-        if predict_days == 60:
-            # 長期預測通常有更大的累積效應
-            predicted_change_pct *= 1.8  # 長期累積效應放大
-        elif predict_days != closest_day:
-            # 對於其他非標準天數，進行線性插值調整
-            adjustment_factor = predict_days / closest_day
-            predicted_change_pct *= adjustment_factor
-        
-        # 【新增】為了兼容性，計算預測價格（僅供參考）
         current_price = latest_data['Close']
         predicted_price = current_price * (1 + predicted_change_pct / 100)
         
-        print(f"XGBoost 預測完成:")
+        print(f"XGBoost 預測結果:")
         print(f"- 預測天數: {predict_days} (使用 {closest_day} 天模型)")
         print(f"- 當前價格: {current_price:.2f}")
         print(f"- 預測漲幅: {predicted_change_pct:+.2f}%")
-        print(f"- 預測價格: {predicted_price:.2f} (參考)")
-        print(f"- 使用特徵數: {len(features_list)} 個")
-        print(f"- 特徵完整性: {completeness:.1f}%")
-        print(f"- 天數調整: {'是' if predict_days != closest_day else '否'}")
+        print(f"- 預測價格: {predicted_price:.2f}")
+        print(f"- 數據時間: {taiex_data.index[-1].strftime('%Y-%m-%d')}")
         
         return {
-            'predicted_price': predicted_price,     # 為了兼容現有代碼
-            'change_pct': predicted_change_pct,     # 【新增】直接的漲幅百分比
-            'confidence': max(0.6, min(0.85, completeness / 100))  # 根據特徵完整性調整信心度
+            'predicted_price': predicted_price,
+            'change_pct': predicted_change_pct,
+            'confidence': 0.75  # 基於模型質量的固定信心度
         }
         
     except Exception as e:
@@ -1120,6 +1102,20 @@ def advanced_xgboost_predict(predict_days=5):
         traceback.print_exc()
         return None
 
+def clear_old_cache():
+    """清理過期的緩存數據"""
+    current_time = time.time()
+    expired_keys = []
+    
+    for key, (data, cache_time) in STOCK_DATA_CACHE.items():
+        if current_time - cache_time > CACHE_EXPIRE_SECONDS:
+            expired_keys.append(key)
+    
+    for key in expired_keys:
+        del STOCK_DATA_CACHE[key]
+        print(f"清理過期緩存: {key}")
+
+
 def get_prediction(data, predict_days=5):
     """
     【【模型預測控制器】】
@@ -1853,45 +1849,66 @@ def update_business_climate_chart(selected_stock):
     [Input('stock-dropdown', 'value'),
      Input('period-dropdown', 'value')]
 )
+ANALYSIS_CACHE = {}
+ANALYSIS_CACHE_DURATION = 60  # 改為1分鐘
+
 def update_analysis_text(selected_stock, period):
-    # 建立快取的唯一鍵值
+    """修改版分析文本更新 - 縮短緩存時間"""
     cache_key = f"{selected_stock}-{period}"
     current_time = time.time()
 
-    # 1. 檢查快取
+    # 檢查緩存
     if cache_key in ANALYSIS_CACHE:
         cached_data = ANALYSIS_CACHE[cache_key]
-        if current_time - cached_data['timestamp'] < CACHE_DURATION_SECONDS:
-            print(f"從快取載入分析: {cache_key}")
-            # 直接回傳快取的內容
+        if current_time - cached_data['timestamp'] < ANALYSIS_CACHE_DURATION:
+            print(f"使用分析緩存: {cache_key} (剩餘: {int(ANALYSIS_CACHE_DURATION - (current_time - cached_data['timestamp']))}秒)")
             return cached_data['technical'], cached_data['fundamental'], cached_data['outlook']
 
     print(f"重新生成分析: {cache_key}")
-    # --- 如果快取沒有，才繼續執行以下程式 ---
-
+    
+    # 清理舊緩存
+    clear_old_cache()
+    
+    # 獲取最新數據
     data = get_stock_data(selected_stock, period)
     stock_name = [name for name, symbol in TAIWAN_STOCKS.items() if symbol == selected_stock][0]
+    
     if data.empty or len(data) < 20:
         return "資料不足，無法分析", "資料不足，無法分析", "資料不足，無法分析"
     
     data = calculate_technical_indicators(data)
     
-    # 2. 技術面分析
+    # 技術面分析（使用最新數據）
     price_change = ((data['Close'].iloc[-1] - data['Close'].iloc[0]) / data['Close'].iloc[0]) * 100
     rsi_current = data['RSI'].iloc[-1] if not pd.isna(data['RSI'].iloc[-1]) else 50
     macd_current = data['MACD'].iloc[-1] if not pd.isna(data['MACD'].iloc[-1]) else 0
     macd_signal_current = data['MACD_Signal'].iloc[-1] if not pd.isna(data['MACD_Signal'].iloc[-1]) else 0
     
+    # 添加數據時間戳信息
+    latest_date = data.index[-1].strftime('%Y-%m-%d')
+    
     technical_text = html.Div([
-        html.P([html.Strong("價格趨勢："), f"在最近 {period} 期間內，{stock_name} 股價呈現", html.Span(f"{'上漲' if price_change > 5 else '下跌' if price_change < -5 else '盤整'}", style={'color': 'red' if price_change > 5 else 'green' if price_change < -5 else 'orange', 'font-weight': 'bold'}), f"走勢，累計變動 {price_change:+.1f}%。"]),
-        html.P([html.Strong("RSI 指標："), f"目前的 RSI 值為 {rsi_current:.1f}，", html.Span("處於超買區(>70)" if rsi_current > 70 else "處於超賣區(<30)" if rsi_current < 30 else "在正常範圍內", style={'color': 'green' if rsi_current > 70 else 'red' if rsi_current < 30 else 'blue', 'font-weight': 'bold'}), "。"]),
-        html.P([html.Strong("MACD 指標："), f"MACD 快線 ({macd_current:.3f}) 目前", html.Span("高於" if macd_current > macd_signal_current else "低於", style={'color': 'red' if macd_current > macd_signal_current else 'green', 'font-weight': 'bold'}), f" Signal 慢線 ({macd_signal_current:.3f})，", f"顯示市場動能偏向{'多頭' if macd_current > macd_signal_current else '空頭'}。"]),
+        html.P([html.Strong("數據時間："), f"{latest_date}"]),
+        html.P([html.Strong("價格趋勢："), f"在最近 {period} 期間內，{stock_name} 股價呈現", 
+                html.Span(f"{'上漲' if price_change > 5 else '下跌' if price_change < -5 else '盤整'}", 
+                         style={'color': 'red' if price_change > 5 else 'green' if price_change < -5 else 'orange', 
+                               'font-weight': 'bold'}), 
+                f"走勢，累計變動 {price_change:+.1f}%。"]),
+        html.P([html.Strong("RSI 指標："), f"目前的 RSI 值為 {rsi_current:.1f}，", 
+                html.Span("處於超買區(>70)" if rsi_current > 70 else "處於超賣區(<30)" if rsi_current < 30 else "在正常範圍內", 
+                         style={'color': 'green' if rsi_current > 70 else 'red' if rsi_current < 30 else 'blue', 
+                               'font-weight': 'bold'}), "。"]),
+        html.P([html.Strong("MACD 指標："), f"MACD 快線 ({macd_current:.3f}) 目前", 
+                html.Span("高於" if macd_current > macd_signal_current else "低於", 
+                         style={'color': 'red' if macd_current > macd_signal_current else 'green', 
+                               'font-weight': 'bold'}), 
+                f" Signal 慢線 ({macd_signal_current:.3f})，顯示市場動能偏向{'多頭' if macd_current > macd_signal_current else '空頭'}。"]),
     ])
     
-    # 3. 基本面與展望分析 (呼叫 Gemini)
+    # AI 分析（基於最新數據）
     fundamental_text, market_outlook_text = generate_gemini_analysis(stock_name, selected_stock, period, data)
     
-    # 4. 將新產生的結果存入快取
+    # 存入緩存
     ANALYSIS_CACHE[cache_key] = {
         'technical': technical_text,
         'fundamental': fundamental_text,
@@ -2360,4 +2377,6 @@ def create_trading_details_table(trades_df):
 
 # 主程式執行
 if __name__ == '__main__':
+    # 初始化應用
+    initialize_app()
     app.run(host="0.0.0.0", port=7860, debug=False)