Update app.py

app.py

@@ -41,25 +41,28 @@ class StockPredictor:
         self.model = None
 
     def preprocess_data(self):
-        # Prophet requires columns named 'ds' and 'y'
         self.data = self.data.reset_index()
         self.data = self.data.rename(columns={'Date': 'ds', 'Close': 'y'})
         
-        # Add any additional features you want to use
-        self.data['SMA_20'] = self.data['y'].rolling(window=20).mean()
-        self.data['EMA_20'] = self.data['y'].ewm(span=20, adjust=False).mean()
-        self.data['RSI'] = self.calculate_rsi(self.data['y'], periods=14)
+        # Add technical indicators
+        self.data['SMA_20'] = SMAIndicator(close=self.data['y'], window=20).sma_indicator()
+        self.data['EMA_20'] = EMAIndicator(close=self.data['y'], window=20).ema_indicator()
+        self.data['RSI'] = RSIIndicator(close=self.data['y'], window=14).rsi()
+        bb = BollingerBands(close=self.data['y'], window=20, window_dev=2)
+        self.data['BB_high'] = bb.bollinger_hband()
+        self.data['BB_low'] = bb.bollinger_lband()
+        
+        # Add lagged features
+        self.data['lag_1'] = self.data['y'].shift(1)
+        self.data['lag_7'] = self.data['y'].shift(7)
+        
+        # Add rolling statistics
+        self.data['rolling_mean_7'] = self.data['y'].rolling(window=7).mean()
+        self.data['rolling_std_7'] = self.data['y'].rolling(window=7).std()
         
         # Handle NaN values
         self.data = self.data.dropna()
 
-    def calculate_rsi(self, prices, periods=14):
-        delta = prices.diff()
-        gain = (delta.where(delta > 0, 0)).rolling(window=periods).mean()
-        loss = (-delta.where(delta < 0, 0)).rolling(window=periods).mean()
-        rs = gain / loss
-        return 100 - (100 / (1 + rs))
-
     def train_model(self):
         try:
             self.model = Prophet(
@@ -72,9 +75,8 @@ class StockPredictor:
             )
             
             # Add additional regressors
-            self.model.add_regressor('SMA_20')
-            self.model.add_regressor('EMA_20')
-            self.model.add_regressor('RSI')
+            for column in ['SMA_20', 'EMA_20', 'RSI', 'BB_high', 'BB_low', 'lag_1', 'lag_7', 'rolling_mean_7', 'rolling_std_7']:
+                self.model.add_regressor(column)
             
             self.model.fit(self.data)
             return True
@@ -87,8 +89,8 @@ class StockPredictor:
             future = self.model.make_future_dataframe(periods=days)
             
             # Add regressor values for future dates
-            for feature in ['SMA_20', 'EMA_20', 'RSI']:
-                future[feature] = self.data[feature].iloc[-1]  # Use last known value
+            for column in ['SMA_20', 'EMA_20', 'RSI', 'BB_high', 'BB_low', 'lag_1', 'lag_7', 'rolling_mean_7', 'rolling_std_7']:
+                future[column] = self.data[column].iloc[-1]  # Use last known value
             
             forecast = self.model.predict(future)
             return forecast
@@ -111,6 +113,11 @@ class StockPredictor:
 
         return mse, mape, rmse
 
+    def cross_validate_model(self):
+        df_cv = cross_validation(self.model, initial='730 days', period='180 days', horizon='90 days')
+        df_p = performance_metrics(df_cv)
+        return df_p
+
 def fetch_stock_data(ticker):
     try:
         end_date = datetime.now()
@@ -180,6 +187,24 @@ def create_prediction_plot(data, predicted_data, company_name):
             name='Predicted Data',
             line=dict(color='yellow')
         ))
+        
+        # Add prediction intervals
+        fig.add_trace(go.Scatter(
+            x=future_dates,
+            y=predicted_data['yhat_upper'],
+            mode='lines',
+            line=dict(width=0),
+            showlegend=False
+        ))
+        fig.add_trace(go.Scatter(
+            x=future_dates,
+            y=predicted_data['yhat_lower'],
+            mode='lines',
+            line=dict(width=0),
+            fillcolor='rgba(255, 255, 0, 0.3)',
+            fill='tonexty',
+            name='Prediction Interval'
+        ))
 
     fig.update_layout(
         title=f'{company_name} Stock Price Prediction',
@@ -246,7 +271,7 @@ def main():
         predict_stock_prices()
 
 def test_model():
-    st.header("Test Prophet Model")
+    st.header("Test Enhanced Prophet Model")
 
     col1, col2 = st.columns(2)
 
@@ -266,13 +291,11 @@ def test_model():
                 st.write(data.describe())
                 st.dataframe(data.head())
 
-                st.markdown(get_table_download_link(data), unsafe_allow_html=True)
-
                 split_index = int(len(data) * (1 - test_split))
                 train_data = data.iloc[:split_index]
                 test_data = data.iloc[split_index:]
 
-                predictor = StockPredictor(train_data)
+                predictor = EnhancedStockPredictor(train_data)
                 predictor.preprocess_data()
                 if predictor.train_model():
                     test_pred = predictor.predict(days=len(test_data))
@@ -290,6 +313,18 @@ def test_model():
 
                             plot = create_test_plot(predictor.data, test_data.reset_index().rename(columns={'Date': 'ds', 'Close': 'y'}), test_pred, company_name)
                             st.plotly_chart(plot, use_container_width=True)
+
+                            # Cross-validation results
+                            st.subheader("Cross-Validation Results")
+                            cv_results = predictor.cross_validate_model()
+                            st.dataframe(cv_results)
+
+                            # Feature importance
+                            st.subheader("Feature Importance")
+                            feature_importance = pd.DataFrame(predictor.model.params['regressor_coefficients'].items(), columns=['Feature', 'Importance'])
+                            feature_importance = feature_importance.sort_values('Importance', ascending=False)
+                            fig = px.bar(feature_importance, x='Feature', y='Importance', title='Feature Importance')
+                            st.plotly_chart(fig, use_container_width=True)
                         else:
                             st.error("Failed to evaluate the model. The evaluation metrics are None.")
                     else:
@@ -298,7 +333,7 @@ def test_model():
                     st.error("Failed to train the Prophet model. Please try a different dataset.")
 
 def predict_stock_prices():
-    st.header("Predict Stock Prices")
+    st.header("Predict Stock Prices with Enhanced Model")
 
     col1, col2 = st.columns(2)
 
@@ -318,9 +353,7 @@ def predict_stock_prices():
                 st.write(data.describe())
                 st.dataframe(data.head())
 
-                st.markdown(get_table_download_link(data), unsafe_allow_html=True)
-
-                predictor = StockPredictor(data)
+                predictor = EnhancedStockPredictor(data)
                 predictor.preprocess_data()
                 if predictor.train_model():
                     predictions = predictor.predict(days=days_to_predict)
@@ -337,6 +370,17 @@ def predict_stock_prices():
                         pred_df.columns = ['Date', 'Predicted Price', 'Lower Bound', 'Upper Bound']
                         st.dataframe(pred_df)
 
+                        # Component-wise forecast
+                        st.subheader("Forecast Components")
+                        fig = predictor.model.plot_components(predictions)
+                        st.pyplot(fig)
+
+                        # Changepoints
+                        st.subheader("Detected Changepoints")
+                        fig = predictor.model.plot(predictions)
+                        a = add_changepoints_to_plot(fig.gca(), predictor.model, predictions)
+                        st.pyplot(fig)
+
                         news = fetch_news(company_name)
                         st.subheader("Latest News")
                         for item in news: