Enhance sentiment analysis in trading strategy within app.py

- Updated get_sentiment method to calculate positive and negative sentiment percentages from news headlines.
- Added handling for cases with no news available, returning None for sentiment data.
- Modified on_trading_iteration to log sentiment data and adjust trading decisions based on new sentiment thresholds.
- Improved trading logic to utilize sentiment percentages instead of binary sentiment classification.

This update refines the trading strategy by incorporating a more nuanced sentiment analysis approach, allowing for better-informed trading decisions.

app.py

@@ -132,21 +132,46 @@ def execute_alpaca_trading():
             three_days_prior = today - Timedelta(days=3)
             return today.strftime('%Y-%m-%d'), three_days_prior.strftime('%Y-%m-%d')
 
-        def get_sentiment(self):
+        def get_sentiment(self): 
             today, three_days_prior = self.get_dates()
-            news = self.api.get_news(symbol=self.symbol, 
-                                     start=three_days_prior, 
-                                     end=today)
-            news = [ev.__dict__["_raw"]["headline"] for ev in news]
-            probability, sentiment = estimate_sentiment(news)
-            return probability, sentiment
+            news = self.api.get_news(symbol=self.symbol, start=three_days_prior, end=today)
+            
+            if not news:
+                return None, None  # No news available
+            
+            # Extract headlines
+            news_headlines = [ev.__dict__["_raw"]["headline"] for ev in news]
+            
+            # Calculate sentiment for each headline
+            positive_count = 0
+            negative_count = 0
+            
+            for headline in news_headlines:
+                probability, sentiment = estimate_sentiment([headline])
+                if sentiment == "positive":
+                    positive_count += 1
+                elif sentiment == "negative":
+                    negative_count += 1
+            
+            total_articles = len(news_headlines)
+            positive_percentage = (positive_count / total_articles) * 100 if total_articles else 0
+            negative_percentage = (negative_count / total_articles) * 100 if total_articles else 0
+
+            return positive_percentage, negative_percentage
+
 
         def on_trading_iteration(self):
-            cash, last_price, quantity = self.position_sizing()
-            probability, sentiment = self.get_sentiment()
+            cash, last_price, quantity = self.position_sizing() 
+            positive_percentage, negative_percentage = self.get_sentiment()
+
+            if positive_percentage is None or negative_percentage is None:
+                self.log("No sentiment data available, skipping this iteration.")
+                return
+
+            self.log(f"Positive Sentiment: {positive_percentage:.2f}%, Negative Sentiment: {negative_percentage:.2f}%")
 
             if cash > last_price:
-                if sentiment == "positive" and probability > .999:
+                if positive_percentage > 60 and negative_percentage < 30:  # Example threshold
                     if self.last_trade == "sell":
                         self.sell_all()
                     order = self.create_order(
@@ -155,11 +180,11 @@ def execute_alpaca_trading():
                         "buy",
                         type="bracket",
                         take_profit_price=last_price * 1.20,
-                        stop_loss_price=last_price * .95
+                        stop_loss_price=last_price * 0.95,
                     )
                     self.submit_order(order)
                     self.last_trade = "buy"
-                elif sentiment == "negative" and probability > .999:
+                elif negative_percentage > 60 and positive_percentage < 30:  # Example threshold
                     if self.last_trade == "buy":
                         self.sell_all()
                     order = self.create_order(
@@ -167,12 +192,13 @@ def execute_alpaca_trading():
                         quantity,
                         "sell",
                         type="bracket",
-                        take_profit_price=last_price * .8,
-                        stop_loss_price=last_price * 1.05
+                        take_profit_price=last_price * 0.8,
+                        stop_loss_price=last_price * 1.05,
                     )
                     self.submit_order(order)
                     self.last_trade = "sell"
 
+
     start_date = datetime(2021, 1, 1)
     end_date = datetime(2024, 10, 1)
     broker = Alpaca(ALPACA_CREDS)