Upload 9 files

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+factor_data.csv filter=lfs diff=lfs merge=lfs -text
+finbert_sentiment.csv filter=lfs diff=lfs merge=lfs -text

app.py

@@ -1,77 +1,255 @@
 # import streamlit as st
 # import pandas as pd
 # import numpy as np
+# import os
+# import ast
+# import openai
+# from openai import OpenAI
+# import json
+# from getpass import getpass
+# from scipy.spatial.distance import cosine
+# from tqdm import tqdm
 # import matplotlib.pyplot as plt
+# import financial_analysis as fa
+# from financial_analysis import alphalens_analysis, alphalens_analysis_by_sector, calculate_information_ratio, process_sentiment_data
 
-# # Placeholder for loading models
-# def load_models():
-#     # In a real scenario, you would load your pre-trained models here.
-#     return {"model_placeholder": "Loaded Model"}
-
-# # Placeholder function to classify news as ESG-related
-# def classify_esg(text, models, api_key):
-#     # Simulate ESG classification logic
-#     # This is where you would use your model to classify the text.
-#     return np.random.choice(["Yes", "No"])
-
-# # Placeholder function to determine sentiment
-# def determine_sentiment(text, models, api_key):
-#     # Simulate sentiment analysis logic
-#     # This is where you would use your model to determine the sentiment.
-#     return np.random.choice(["Positive", "Neutral", "Negative"])
-
-# # Placeholder function to run Alphalens analysis
-# def run_alphalens_analysis(data, models, api_key):
-#     # Simulate some metrics
-#     metrics = {"alpha": np.random.rand(), "beta": np.random.rand()}
-    
-#     # Generate a simple plot
-#     fig, ax = plt.subplots()
-#     ax.plot([1, 2, 3], [1, 2, 3], 'r')  # Example plot
-#     ax.set_title('Example Plot')
-    
-#     return metrics, [fig]
 
-# # Streamlit app code
-# models = load_models()
+# def get_sentiment_gpt(company, SASB, news, max_retries=5, model = 'gpt-4-turbo-2024-04-09'):
+#     system_prompt = """
+#     As a specialist in ESG analytics, 
+#     You possess a deep understanding of evaluating environmental, social, and governance factors in the context of corporate news. 
+#     Your expertise lies in discerning the underlying sentiment of news segments that pertain to a company's ESG practices, 
+#     determining whether the coverage reflects a positive, negative, or neutral stance.
+#     """
 
-# st.title('NLP Project: ESG News Analysis and Financial Impact')
+#     allowed_sentiments = ['Negative', 'Positive', 'Neutral']
+#     attempt = 0
 
-# api_key = st.sidebar.text_input("OpenAI API Key", type="password")
+#     while attempt < max_retries:
+#         main_prompt = f"""
+#         Classify the sentiment (Only options: Positive, Negative, Neutral) of the following news: {news} |
+#         The sentiment classification should be about the sections of the news talking about the company {company}. | 
+#         The ESG part of the news should be around topics within the following SASB topics {SASB}
 
-# uploaded_file = st.file_uploader("Choose a CSV file", type="csv")
-# if uploaded_file is not None:
-#     data = pd.read_csv(uploaded_file)
-#     st.write("Uploaded News Data:")
-#     st.dataframe(data)
+#         The output should be a structured JSON object with the key: "sentiment".
 
-#     if st.button('Classify News as ESG-related'):
-#         data['ESG'] = data['news'].apply(lambda x: classify_esg(x, models, api_key))
-#         st.write("News with ESG Classification:")
-#         st.dataframe(data)
+#         Here is the format I expect for the JSON object:
 
-#     if st.button('Determine Sentiment'):
-#         data['Sentiment'] = data['news'].apply(lambda x: determine_sentiment(x, models, api_key))
-#         st.write("News with Sentiment Analysis:")
-#         st.dataframe(data)
+#         {{
+#           "sentiment": "Enter 'Positive', 'Neutral', or 'Negative'",
+#         }}
 
-#     if st.button('Run Alphalens Analysis'):
-#         metrics, plots = run_alphalens_analysis(data, models, api_key)
-#         st.write("Alphalens Analysis Metrics:")
-#         st.json(metrics)
-        
-#         st.write("Alphalens Analysis Plots:")
-#         for plot in plots:
-#             st.pyplot(plot)
+#         Do not return any additional text or information outside of this JSON structure.
+#         """
+
+#         messages = [
+#           {"role": "system", "content": system_prompt},
+#           {"role": "user", "content": main_prompt}
+#         ]
+
+#         response = openai.chat.completions.create(
+#         model=model,
+#         messages=messages,
+#         response_format={"type": "json_object"}  # Enable JSON mode
+#         )
+
+#         response_json = json.loads(response.choices[0].message.content)
+#         json_sentiment = response_json.get('sentiment')
+
+#         if json_sentiment in allowed_sentiments:
+#             return json_sentiment
 
+#         attempt += 1
+
+#     # After max retries, if no valid sentiment is found, handle as needed (e.g., return a default sentiment)
+#     print("Failed to obtain a valid sentiment after maximum retries. Defaulting to 'Neutral'.")
+#     return 'Neutral'  # Default return value if no valid sentiment is obtained
+
+
+# def update_dataset_with_gpt_sentiment(df, model, column_name='GPT_based_sentiment'):
+#     # Initialize the new column to store GPT-based sentiment
+#     df['GPT_based_sentiment'] = None
+    
+#     # Use tqdm to show a progress bar for the operation
+#     for index, row in tqdm(df.iterrows(), total=len(df), desc="Processing rows"):
+#         # Extract necessary information for each row
+#         company = row['Company']  # Make sure this matches your DataFrame's column name
+#         SASB = row['SASB']  # Make sure this matches your DataFrame's column name
+#         news = row['title & content']  # Make sure this matches your DataFrame's column name
+        
+#         # Call the function to get the sentiment
+#         sentiment = get_sentiment_gpt(company, SASB, news, model=model)
+        
+#         # Update the DataFrame with the obtained sentiment
+#         df.at[index, column_name] = sentiment  # Now correctly assigns the sentiment
+        
+#     return df
+
+# # Function to get embeddings, provided by you
+# def get_embedding(text, model="text-embedding-3-small"):
+#    client = OpenAI()
+#    text = text.replace("\n", " ")
+#    return client.embeddings.create(input=[text], model=model).data[0].embedding
+
+# # Function to calculate cosine similarity
+# def cosine_similarity(v1, v2):
+#     return 1 - cosine(v1, v2)
+
+# def calculate_sasb_embeddings(sasb_str):
+#     # Safely convert the string representation of a dictionary into an actual dictionary
+#     try:
+#         sasb_dict = ast.literal_eval(sasb_str)
+#         if not isinstance(sasb_dict, dict):
+#             raise ValueError("SASB column does not contain a valid dictionary.")
+#     except ValueError as e:
+#         print(f"Error converting SASB column to dictionary: {e}")
+#         return {}
+    
+#     sasb_embeddings = {}
+#     for topic, content in sasb_dict.items():
+#         # Join the list of keywords into a single string
+#         combined_content = ' '.join(content)
+#         sasb_embeddings[topic] = get_embedding(combined_content)
+#     return sasb_embeddings
+
+# # Function to process ESG classification
+# def classify_esg(data):
+#     # Calculate embeddings for the news
+#     data['news_embeddings'] = data['title & content'].apply(get_embedding)
+
+#     # Calculate embeddings for SASB topics (you need to have your SASB topics defined)
+#     data['sasb_embeddings'] = data['SASB'].apply(calculate_sasb_embeddings)
+
+#     # Compute cosine similarities
+#     data['cosine_similarities'] = data.apply(
+#         lambda row: {topic: cosine_similarity(row['news_embeddings'], emb)
+#                      for topic, emb in row['sasb_embeddings'].items()},
+#         axis=1
+#     )
+
+#     # Extract max cosine similarity
+#     data['max_cosine_similarity'] = data['cosine_similarities'].apply(lambda x: max(x.values()))
+
+#     # Mark the top 10% of news by max_cosine_similarity within each 'Sector' as 'Yes'
+#     sector_thresholds = data.groupby('Sector')['max_cosine_similarity'].quantile(0.9).to_dict()
+#     data['ESG_relevance'] = data.apply(
+#         lambda row: 'Yes' if row['max_cosine_similarity'] >= sector_thresholds[row['Sector']] else 'No',
+#         axis=1
+#     )
+#     return data
+
+# def main():
+#     st.set_page_config(page_title="NLP ESG Project", page_icon="📈")
+
+#     # Custom styles
+#     st.markdown(
+#         """
+#         <style>
+#         .streamlit-container {
+#             background-color: #F5F5F5;
+#         }
+#         .stButton>button {
+#             width: 100%;
+#             border-radius: 10px;
+#             border: none;
+#             margin: 10px 0;
+#             padding: 15px 20px;
+#             background-color: #79AEC8;
+#             color: white;
+#             font-size: 18px;
+#         }
+#         .stButton>button:hover {
+#             background-color: #6699CC;
+#         }
+#         </style>
+#         """,
+#         unsafe_allow_html=True,
+#     )
+
+#     # Header section
+#     st.write("# NLP Project: ESG News Analysis and Financial Impact")
+#     st.sidebar.write("## Configuration")
+
+#     # API Key input
+#     openai_api_key = st.sidebar.text_input("Enter your OpenAI API key", type="password")
+#     openai_api_key = os.getenv('OPENAI_API_KEY')
+#     os.environ["OPENAI_API_KEY"] = openai_api_key
+#     openai.api_key = openai_api_key
+
+#     # File Upload
+#     st.sidebar.write("## Upload Data")
+#     uploaded_file = st.sidebar.file_uploader("", type="csv")
+
+#     # Investment Strategy Slider
+#     st.sidebar.markdown("### Investment Strategy")
+#     investment_strategy = st.sidebar.slider(
+#         "Investment Strategy",
+#         min_value=0.0, max_value=1.0, value=0.5, step=0.01,
+#         format="",
+#         help="0 is Conservative, 1 is Aggressive",
+#         label_visibility="collapsed"
+#     )
+#     st.sidebar.text(f"Current Strategy: {'Conservative' if investment_strategy <= 0.5 else 'Aggressive'}")
+
+#     # Main container
+
+#     st.sidebar.write("## Upload Data")
+#     uploaded_file = st.sidebar.file_uploader("Please upload a CSV file", type="csv", label_visibility="collapsed")
+
+#     if uploaded_file:
+#         # Displaying the file
+#         data = pd.read_csv(uploaded_file)
+#         st.session_state.classified_data = classify_esg(data)
+#         st.write("### Uploaded News Data:")
+#         st.dataframe(data, use_container_width=True)
+
+#         if st.button("🔍 Classify ESG"):
+#             st.write("Classifying ESG-related news...")
+#             try:
+#                 with st.spinner("Calculating embeddings and similarities..."):
+#                     st.session_state.classified_data = classify_esg(st.session_state.classified_data)
+#                     st.write("Classified News Data:")
+#                     st.dataframe(st.session_state.classified_data, use_container_width=True)
+#             except Exception as e:
+#                 st.error(f"An error occurred: {e}")
+
+#         if st.button("😊 Determine Sentiment"):
+#             st.write("Determining sentiment using GPT...")
+#             # Run sentiment analysis with GPT
+#             try:
+#                 with st.spinner("Analyzing sentiment..."):
+#                     # Assume you have your API key set and a function defined to handle sentiment analysis
+#                     st.session_state.updated_data = update_dataset_with_gpt_sentiment(st.session_state.classified_data, model='gpt-4-turbo-2024-04-09')
+#                     st.write("News with GPT-based Sentiment Analysis:")
+#                     st.dataframe(st.session_state.updated_data, use_container_width=True)
+#             except Exception as e:
+#                 st.error(f"An error occurred: {e}")
+
+#         if st.button("📊 Alphalens Analysis"):
+#             # process_sentiment_data(sentiment_data = 'finbert_sentiment.csv', sector_ticker = 'sector_ticker.csv', prices = 'prices.csv')
+#             prices = pd.read_csv('prices.csv')
+#             factor_data = pd.read_csv('factor_data.csv')
+#             merged_data = pd.read_csv('merged_data.csv')
+
+#             alphalens_analysis(merged_data, prices)
+
+#         # Expander for advanced settings
+#         with st.expander("Advanced Settings"):
+#             st.write("Any advanced settings and configurations will go here.")
+
+# if __name__ == "__main__":
+#     main()
 
 import streamlit as st
 import pandas as pd
 import numpy as np
 import os
+import ast
 import openai
+from openai import OpenAI
 import json
 from getpass import getpass
+from scipy.spatial.distance import cosine
 from tqdm import tqdm
 import matplotlib.pyplot as plt
 
@@ -146,7 +324,60 @@ def update_dataset_with_gpt_sentiment(df, model, column_name='GPT_based_sentimen
         
     return df
 
-def app_layout():
+# Function to get embeddings, provided by you
+def get_embedding(text, model="text-embedding-3-small"):
+   client = OpenAI()
+   text = text.replace("\n", " ")
+   return client.embeddings.create(input=[text], model=model).data[0].embedding
+
+# Function to calculate cosine similarity
+def cosine_similarity(v1, v2):
+    return 1 - cosine(v1, v2)
+
+def calculate_sasb_embeddings(sasb_str):
+    # Safely convert the string representation of a dictionary into an actual dictionary
+    try:
+        sasb_dict = ast.literal_eval(sasb_str)
+        if not isinstance(sasb_dict, dict):
+            raise ValueError("SASB column does not contain a valid dictionary.")
+    except ValueError as e:
+        print(f"Error converting SASB column to dictionary: {e}")
+        return {}
+    
+    sasb_embeddings = {}
+    for topic, content in sasb_dict.items():
+        # Join the list of keywords into a single string
+        combined_content = ' '.join(content)
+        sasb_embeddings[topic] = get_embedding(combined_content)
+    return sasb_embeddings
+
+# Function to process ESG classification
+def classify_esg(data):
+    # Calculate embeddings for the news
+    data['news_embeddings'] = data['title & content'].apply(get_embedding)
+
+    # Calculate embeddings for SASB topics (you need to have your SASB topics defined)
+    data['sasb_embeddings'] = data['SASB'].apply(calculate_sasb_embeddings)
+
+    # Compute cosine similarities
+    data['cosine_similarities'] = data.apply(
+        lambda row: {topic: cosine_similarity(row['news_embeddings'], emb)
+                     for topic, emb in row['sasb_embeddings'].items()},
+        axis=1
+    )
+
+    # Extract max cosine similarity
+    data['max_cosine_similarity'] = data['cosine_similarities'].apply(lambda x: max(x.values()))
+
+    # Mark the top 10% of news by max_cosine_similarity within each 'Sector' as 'Yes'
+    sector_thresholds = data.groupby('Sector')['max_cosine_similarity'].quantile(0.9).to_dict()
+    data['ESG_relevance'] = data.apply(
+        lambda row: 'Yes' if row['max_cosine_similarity'] >= sector_thresholds[row['Sector']] else 'No',
+        axis=1
+    )
+    return data
+
+def main():
     st.set_page_config(page_title="NLP ESG Project", page_icon="📈")
 
     # Custom styles
@@ -180,6 +411,7 @@ def app_layout():
 
     # API Key input
     openai_api_key = st.sidebar.text_input("Enter your OpenAI API key", type="password")
+    openai_api_key = os.getenv('OPENAI_API_KEY')
     os.environ["OPENAI_API_KEY"] = openai_api_key
     openai.api_key = openai_api_key
 
@@ -199,16 +431,22 @@ def app_layout():
     st.sidebar.text(f"Current Strategy: {'Conservative' if investment_strategy <= 0.5 else 'Aggressive'}")
 
     # Main container
-    if uploaded_file is not None:
+    if uploaded_file:
         # Displaying the file
         data = pd.read_csv(uploaded_file)
+        st.session_state.classified_data = classify_esg(data)
         st.write("### Uploaded News Data:")
         st.dataframe(data, use_container_width=True)
 
         if st.button("🔍 Classify ESG"):
             st.write("Classifying ESG-related news...")
-            # Placeholder - replace with actual ESG classification code
-            data['ESG'] = "Yes"  # placeholder
+            try:
+                with st.spinner("Calculating embeddings and similarities..."):
+                    st.session_state.classified_data = classify_esg(st.session_state.classified_data)
+                    st.write("Classified News Data:")
+                    st.dataframe(st.session_state.classified_data, use_container_width=True)
+            except Exception as e:
+                st.error(f"An error occurred: {e}")
 
         if st.button("😊 Determine Sentiment"):
             st.write("Determining sentiment using GPT...")
@@ -216,9 +454,9 @@ def app_layout():
             try:
                 with st.spinner("Analyzing sentiment..."):
                     # Assume you have your API key set and a function defined to handle sentiment analysis
-                    updated_data = update_dataset_with_gpt_sentiment(data, model='gpt-4-turbo-2024-04-09')
+                    st.session_state.updated_data = update_dataset_with_gpt_sentiment(st.session_state.classified_data, model='gpt-4-turbo-2024-04-09')
                     st.write("News with GPT-based Sentiment Analysis:")
-                    st.dataframe(updated_data, use_container_width=True)
+                    st.dataframe(st.session_state.updated_data, use_container_width=True)
             except Exception as e:
                 st.error(f"An error occurred: {e}")
 
@@ -229,8 +467,5 @@ def app_layout():
         with st.expander("Advanced Settings"):
             st.write("Any advanced settings and configurations will go here.")
 
-def main():
-    app_layout()
-
 if __name__ == "__main__":
-    main()
+    main()

equal_weighted_benchmark_df.csv

@@ -0,0 +1,249 @@
+,date,equal_weighted_benchmark,5d_mean_return,10d_mean_return,20d_mean_return
+21,2022-09-14 00:00:00+00:00,-0.003676159000736702,-0.004098304655096071,-0.008331580823424601,-0.0043233482313036415
+22,2022-09-15 00:00:00+00:00,-0.014448647849191252,-0.00811570954039362,-0.007425553791892592,-0.005290966130099582
+23,2022-09-16 00:00:00+00:00,-0.01018150984258006,-0.010291577429425948,-0.007172411076246634,-0.0023074253300825035
+24,2022-09-19 00:00:00+00:00,0.009131715240025375,-0.010138971898947535,-0.006049274613087459,-0.0023989673424125372
+25,2022-09-20 00:00:00+00:00,-0.0013076505427114021,-0.011908672204641485,-0.004016785108368158,-0.001667495719771195
+26,2022-09-21 00:00:00+00:00,-0.02410392747782434,-0.012843596836909185,-0.0033574809954569616,-0.0022232276080379823
+27,2022-09-22 00:00:00+00:00,-0.025717711927302645,-0.007054626047366221,-0.0005691883822669203,-0.0012626184643332202
+28,2022-09-23 00:00:00+00:00,-0.00921157187938197,-0.004281643476182188,0.0003177701662729227,-0.0005353617907464471
+29,2022-09-26 00:00:00+00:00,-0.00048038023989492915,-0.0020482297281117206,0.00022344205284363693,0.0014588809659362722
+30,2022-09-27 00:00:00+00:00,-0.006307341359481893,0.004207864879698234,-0.001060793557155493,0.0017115541097231192
+31,2022-09-28 00:00:00+00:00,0.006330843278072321,0.006711352378629434,-0.00029325822628145236,0.0032371122185726256
+32,2022-09-29 00:00:00+00:00,-0.011724517037708796,0.00620640676792534,-0.003413101647874253,0.0032347424556726633
+33,2022-09-30 00:00:00+00:00,0.0018495754567621427,0.004946832290400967,0.0027578805195729995,0.0036136359005584486
+34,2022-10-03 00:00:00+00:00,0.031018002364648946,0.002382947233928605,0.0013140047327894195,0.004264118947733028
+35,2022-10-04 00:00:00+00:00,0.00580128834931611,-0.006320930191630265,0.0007028635860197298,0.0031912776861333127
+36,2022-10-05 00:00:00+00:00,0.0038416563336735906,-0.007159056539681236,-0.0011526691448603224,0.002274586532981925
+37,2022-10-06 00:00:00+00:00,-0.01798577177022608,-0.01268015834390541,-0.0019770040155771347,0.00042830915460623605
+38,2022-10-07 00:00:00+00:00,-0.0106111914983441,0.0005950230744271944,-0.0013664621449848522,0.0025698646510722016
+39,2022-10-10 00:00:00+00:00,-0.012968828756671921,0.00028552263376673383,0.002645817758450723,0.003383232250678209
+40,2022-10-11 00:00:00+00:00,0.0015507390715413976,0.007995292527418242,0.0044506647980164125,0.004404850114552215
+41,2022-10-12 00:00:00+00:00,-0.025093912254979374,0.0050514656435360515,0.006740026005053216,0.004128657804646016
+42,2022-10-13 00:00:00+00:00,0.05200138806905839,0.009344607857668694,0.010223197273261998,0.006801831145217036
+43,2022-10-14 00:00:00+00:00,-0.012209288945237099,-0.0033260930873431956,0.00427888322523018,0.005193443672798097
+44,2022-10-17 00:00:00+00:00,0.025213559469038502,0.005023441177708918,0.007086754114892575,0.006035676422397556
+45,2022-10-18 00:00:00+00:00,-0.012778164543042156,0.0008588121267430121,0.0055942624511079436,0.004999830632095196
+46,2022-10-19 00:00:00+00:00,-0.004593051792337033,0.008194656122117549,0.005763452950807643,0.005127444102465934
+47,2022-10-20 00:00:00+00:00,-0.011707362492000929,0.010570150789625577,0.002867315957998419,0.004374917388519577
+48,2022-10-21 00:00:00+00:00,0.02959810000422721,0.01211361422917934,0.006601291022304565,0.006118164276748382
+49,2022-10-24 00:00:00+00:00,0.004351857113112529,0.008985688957480361,0.004032491879711047,0.004075057070100423
+50,2022-10-25 00:00:00+00:00,0.02353960069183561,0.01029924636627716,0.004169854287116967,0.004362253021904609
+51,2022-10-26 00:00:00+00:00,0.006655722595231783,0.0031780048609374107,0.0014094625537276857,0.003422747853239372
+52,2022-10-27 00:00:00+00:00,-0.0045738265529560165,-0.0045362994782050204,0.0030547040125687365,0.0033921065755376425
+53,2022-10-28 00:00:00+00:00,0.014491488053761043,0.0009977835965575294,0.00584218970073483,0.0033196750893709414
+54,2022-10-31 00:00:00+00:00,0.010623521518342454,-0.0008957028827467486,0.0046466853676672415,0.002123153162326777
+55,2022-11-01 00:00:00+00:00,-0.011149948426442515,-0.0019018448298914236,0.00417378663201427,0.0017436676511514526
+56,2022-11-02 00:00:00+00:00,-0.0317018465087496,-0.00038511181416520335,0.004241303048243899,0.0037937682687762425
+57,2022-11-03 00:00:00+00:00,0.023672653813776284,0.010907415734302213,0.005717489711766106,0.004667984801179622
+58,2022-11-04 00:00:00+00:00,0.004877327738659193,0.010726166608973907,0.0053118296322522945,0.0035912822399911896
+59,2022-11-07 00:00:00+00:00,0.0056193648553276125,0.010365722373878391,0.003976180662520567,0.0026974750636089826
+60,2022-11-08 00:00:00+00:00,-0.0034980949721803317,0.010454171748779021,0.004389242344400509,0.0017444841092120649
+61,2022-11-09 00:00:00+00:00,0.023096640255727902,0.008962463760018928,0.005381134793118519,0.0022076431310965885
+62,2022-11-10 00:00:00+00:00,0.022445370157588133,0.0004473559719162085,0.0036342320147014326,0.0008981438759837245
+63,2022-11-11 00:00:00+00:00,0.0029647455307566235,-0.00010862140713379332,0.0007734572539003317,-0.0003605695447114842
+64,2022-11-14 00:00:00+00:00,0.006009028155624304,-0.002092974244106969,-0.00027674591649861757,0.0013991941478267607
+65,2022-11-15 00:00:00+00:00,-0.010367358114339606,-0.0013213151032776165,-0.00054064586546945,0.00010610733776163811
+66,2022-11-16 00:00:00+00:00,-0.01783699590204662,0.001897438852983208,0.003294878175626652,-0.00027251094978617414
+67,2022-11-17 00:00:00+00:00,0.019555673624933614,0.006828897752946318,0.0034445923547001493,-0.0003571100874945935
+68,2022-11-18 00:00:00+00:00,-0.0072918795158902795,0.001657995041880389,0.0018145154577246846,-0.0013897707927707614
+69,2022-11-21 00:00:00+00:00,0.009723643434801197,0.0015510603620931208,0.0013430026592687863,-0.0015683766710776707
+70,2022-11-22 00:00:00+00:00,0.005759780551965727,0.0002711357374123586,-0.0008967914886358623,-0.001495973072915131
+71,2022-11-23 00:00:00+00:00,0.006295096012988876,0.0046833577595701125,-0.0009277130177343687,-0.0017999783977110737
+72,2022-11-25 00:00:00+00:00,-0.005976146401571202,8.099722632682047e-05,-0.0017781511474425098,-0.0022840369626763724
+73,2022-11-28 00:00:00+00:00,-0.007814302311672312,0.001980780515448426,-0.0015152294865902847,-0.0015899960613449957
+74,2022-11-29 00:00:00+00:00,0.0032986264786386797,0.0011151282248217215,0.0030806208868460055,-0.001217022491572957
+75,2022-11-30 00:00:00+00:00,0.0279746941884447,-0.002047443340334215,0.0007466396006418766,-0.0017768779636501985
+76,2022-12-01 00:00:00+00:00,-0.016352882788051644,-0.006394598044119795,-0.0037531595621781923,-0.002941462762931817
+77,2022-12-02 00:00:00+00:00,0.00337744659239721,-0.003659860893787665,-0.004035585947964563,-0.001753146786279652
+78,2022-12-05 00:00:00+00:00,-0.01209042554527834,-0.004920695037299253,-0.004520806783547946,-0.0019170326401667988
+79,2022-12-06 00:00:00+00:00,-0.01260211701899184,0.005008673283487237,-0.004411106239523519,-0.0012863211194042838
+80,2022-12-07 00:00:00+00:00,0.0054668424990757655,0.003578888247210838,-0.0021108389165162425,-0.00039497903697917883
+81,2022-12-08 00:00:00+00:00,-0.0025281110944274376,-0.001158953755318699,-0.002712652494580931,0.0002611377891841786
+82,2022-12-09 00:00:00+00:00,-0.0030581677096146818,-0.004485304011963333,-0.002838439871711985,0.00012013770303957311
+83,2022-12-12 00:00:00+00:00,0.038459837053426336,-0.004204296782894843,-0.001682117639733388,0.000955921367436175
+84,2022-12-13 00:00:00+00:00,-0.019439876506603074,-0.013443002800764471,-0.005299096662726175,-0.00042983011041462674
+85,2022-12-14 00:00:00+00:00,-0.017943709493747392,-0.007634336469754126,-0.0042559273519537375,0.0001271483306378504
+86,2022-12-15 00:00:00+00:00,-0.019156928642757205,-0.004258643551772655,-0.002209615432008975,0.0016259677434219133
+87,2022-12-16 00:00:00+00:00,-0.0016701182490717148,-0.0012211795063897244,0.0005605489276259438,0.0026765297699224036
+88,2022-12-19 00:00:00+00:00,-0.010985076265142509,0.0008081326010809642,0.0007033595870670358,0.0013818632373116411
+89,2022-12-20 00:00:00+00:00,0.011286800676776778,0.0029639323447359475,0.001934726087374865,0.001938090675887691
+90,2022-12-21 00:00:00+00:00,-0.0007796221147050135,-0.0009578917211637821,0.00135614882488803,0.002144071681145062
+91,2022-12-22 00:00:00+00:00,-0.0038849057168939783,-0.00015662701462072144,0.003366475552433948,0.002499192033871605
+92,2022-12-23 00:00:00+00:00,0.008522759321651276,0.002373848466957648,0.0032046714130471816,0.002343718221016368
+93,2022-12-27 00:00:00+00:00,-0.0004069620633721189,0.000606894125618909,0.003732794819247755,0.0027976522861079722
+94,2022-12-28 00:00:00+00:00,-0.008134940624643141,0.0009048488555181238,0.004887554807300264,0.002836614455128162
+95,2022-12-29 00:00:00+00:00,0.003138254119539179,0.003677972424751826,0.004841251866237464,0.0031621774740277484
+96,2022-12-30 00:00:00+00:00,0.008862472869208805,0.006892494230496532,0.005684334529429424,0.0029175460267830363
+97,2023-01-03 00:00:00+00:00,-0.0004454056107618189,0.00393891621987011,0.004802254891621128,0.002913869213491017
+98,2023-01-04 00:00:00+00:00,0.001066202344047466,0.0068204956456235834,0.0021124611665669065,0.003965826277727658
+99,2023-01-05 00:00:00+00:00,0.00539287954472177,0.008744123929100633,0.0019087545284069858,0.0035962825339874486
+100,2023-01-06 00:00:00+00:00,0.019080784087096292,0.0058348405945136605,0.002919496869240496,0.0029248423599723596
+101,2023-01-09 00:00:00+00:00,-0.005423323975577465,0.004225346680399479,0.0015359724910240552,0.0017124357670201382
+102,2023-01-10 00:00:00+00:00,0.013422394452403931,0.005490505891646507,0.0013807969121424908,0.002336908981798623
+103,2023-01-11 00:00:00+00:00,0.010501782665511043,-0.0025566610321382474,0.0017008707949891402,0.0017691969049688369
+104,2023-01-12 00:00:00+00:00,-0.008683084200997962,-0.004771227639351861,0.0006392084677149688,0.00024085982014168195
+105,2023-01-13 00:00:00+00:00,0.011211035819871068,-6.135031418093928e-05,0.001331736307231877,0.0010561955271994425
+106,2023-01-17 00:00:00+00:00,0.0005965124072326647,-0.001174437672390838,6.623370925527113e-05,0.00078895049489153
+107,2023-01-18 00:00:00+00:00,-0.025959436629375873,-0.002652241101487527,0.0009280279345257354,0.000520682708660935
+108,2023-01-19 00:00:00+00:00,-0.0007792090132950034,0.005998389613138973,0.005556821803484369,0.001790337226031672
+109,2023-01-20 00:00:00+00:00,0.015171332460758516,0.0061634913246250735,0.005084514674461852,0.001694383648083702
+110,2023-01-23 00:00:00+00:00,0.005542971074545104,0.0026606667064368937,0.002725384864575145,0.0005829093622090459
+111,2023-01-24 00:00:00+00:00,-0.006622727888738723,0.0013327048350599657,0.0017756832880997334,-0.0002611183863901375
+112,2023-01-25 00:00:00+00:00,0.016496864760670086,0.00456974537899959,0.0031524633433066183,0.00027312040798949794
+113,2023-01-26 00:00:00+00:00,8.113506539531873e-05,0.004926332892019173,0.001729924350080279,-0.0012505797471628595
+114,2023-01-27 00:00:00+00:00,-0.0019495737438668711,0.003857271982804954,-0.00019183771654848798,-0.0006585410819113961
+115,2023-01-30 00:00:00+00:00,-0.00099148612495581,0.002715162306309398,0.0007423704742227855,-0.0009786871366024893
+116,2023-01-31 00:00:00+00:00,0.009280402583510174,0.0021854906270825964,0.0014632424254074508,-0.0011627945038930286
+117,2023-02-01 00:00:00+00:00,0.01807208776262952,0.001624306926245983,5.5996165418699724e-05,-0.001850062269103672
+118,2023-02-02 00:00:00+00:00,-0.005286271143741116,-0.0014152827676169523,-0.0018714186626452999,-0.0018426042323073531
+119,2023-02-03 00:00:00+00:00,-0.00735610191660725,-0.004034184361960209,-0.0015952112175047963,-0.0011180493955347387
+120,2023-02-06 00:00:00+00:00,-0.0035404118498477645,-0.0010880312237595148,-0.0014860785350897109,-0.0010126401388516049
+121,2023-02-07 00:00:00+00:00,0.006591189675830193,0.0008091400921922269,-0.002227611292490897,-0.0016015572664364727
+122,2023-02-08 00:00:00+00:00,0.002722672074499014,-0.00145984913090635,-0.002481905827220295,-0.001753163968636046
+123,2023-02-09 00:00:00+00:00,-0.01863649800814907,-0.002328582952201085,-0.00411753055898942,-0.0030074135500991186
+124,2023-02-10 00:00:00+00:00,0.007678079567730095,0.0009154827917578608,-0.0011238194828288672,-0.0035726306795227088
+125,2023-02-13 00:00:00+00:00,0.006112763508391236,-0.0018560342559001501,-0.0026927013376944766,-0.0029216814590479244
+126,2023-02-14 00:00:00+00:00,-0.00477608908294868,-0.005254346119167877,-0.0037429459423787784,-0.004198386147281157
+127,2023-02-15 00:00:00+00:00,-0.0016779104300324093,-0.003538787850179554,-0.0037421445979662895,-0.004128963605615242
+128,2023-02-16 00:00:00+00:00,-0.0025715184411038345,-0.0059761742312147694,-0.0018586028199844477,-0.0033253898156173634
+129,2023-02-17 00:00:00+00:00,-0.006241038440088623,-0.0031328292440078054,-0.000669474121235514,-0.003639933277377052
+130,2023-02-21 00:00:00+00:00,-0.011079628060444846,-0.003524099040014799,-0.0005459995477119063,-0.002429861328467912
+131,2023-02-22 00:00:00+00:00,0.003989484205406339,-0.0022725958242575377,-0.0009942324563652996,-0.0019073630129376826
+132,2023-02-23 00:00:00+00:00,-0.014204582469811391,-0.004012438705374892,-0.0010596116561342839,-0.002917469152075522
+133,2023-02-24 00:00:00+00:00,0.012116901005240929,0.0023769684864330125,-0.001985397748864818,-0.0028627172299213795
+134,2023-02-27 00:00:00+00:00,-0.008285144803713385,0.0018614725168118733,-0.006104966124183615,-0.0022968902683756696
+135,2023-02-28 00:00:00+00:00,-0.004807402525172574,0.0024761599269984085,-0.0032439344630809767,-0.0020429450955181806
+136,2023-03-01 00:00:00+00:00,-0.004827232807062231,0.00027420356036656233,-0.00485372596501946,-0.0011934862887619702
+137,2023-03-02 00:00:00+00:00,0.017930193919978674,0.0019492770509415744,-0.004661128176875345,-0.0003157692613777001
+138,2023-03-03 00:00:00+00:00,0.009529409984429534,-0.00627793919095608,-0.0048926759315512025,-0.0011570698350831749
+139,2023-03-06 00:00:00+00:00,-0.005176985690555578,-0.013935513609302596,-0.006659345106030147,-0.0010800877189296768
+140,2023-03-07 00:00:00+00:00,-0.015659807800211298,-0.008892647319852509,-0.004347146058003588,-0.0007134469206721708
+141,2023-03-08 00:00:00+00:00,0.003381326477880937,-0.010040088163365053,-0.0028445778739169805,-0.0006077551608841233
+142,2023-03-09 00:00:00+00:00,-0.023663304953076534,-0.01129818982215423,-0.00485740218875114,-0.0007560598528678973
+143,2023-03-10 00:00:00+00:00,-0.030893398613373615,-0.0038206077045881684,-0.0039120160508721415,0.0002554549397418739
+144,2023-03-13 00:00:00+00:00,0.022362175123981063,0.0004726877276335097,0.0015302959531613201,0.0025145261615524264
+145,2023-03-14 00:00:00+00:00,-0.022348761476505004,3.87366713232587e-05,-0.0009854352607259823,0.001845601878591786
+146,2023-03-15 00:00:00+00:00,-0.003325296700711762,0.0046958963994744455,0.0025982415691328847,0.002495461635709717
+147,2023-03-16 00:00:00+00:00,0.015840822248179467,0.0018117355447859551,0.004335236186220326,0.002976216305923911
+148,2023-03-17 00:00:00+00:00,-0.00946384581669036,-0.004157422162841717,0.002765511695353645,0.0019509949740394807
+149,2023-03-20 00:00:00+00:00,0.020467800118520695,0.002622247225075212,0.0049234985025505,0.0029354726394914344
+150,2023-03-21 00:00:00+00:00,0.0003277948707486952,-0.0020523701485808234,0.0030561322119619866,0.001582701834217248
+151,2023-03-22 00:00:00+00:00,-0.017672005575669312,0.0005096764402386216,0.001572604648531574,0.00149770455947494
+152,2023-03-23 00:00:00+00:00,-0.014087653674746592,0.006824496724776386,0.003478823155561836,0.0025724538644252156
+153,2023-03-24 00:00:00+00:00,0.02506545690589547,0.00997544984123682,0.0046701335408740295,0.0032502924928142535
+154,2023-03-27 00:00:00+00:00,-0.003077478717735781,0.007153683367903214,0.0034858729391427554,0.0017333852376485645
+155,2023-03-28 00:00:00+00:00,0.013277255588550533,0.008286840846777166,0.004760086452127288,0.0011016864749227807
+156,2023-03-29 00:00:00+00:00,0.01328306158951853,0.0026215148784006865,0.0023383992050122837,0.0002527685478874206
+157,2023-03-30 00:00:00+00:00,0.0007077524431417956,0.00012008894023344704,0.0015562642179077924,0.0001643135986638564
+158,2023-03-31 00:00:00+00:00,0.011213629538062238,-0.0006085838330772233,0.0010980689017193001,0.0007130687983601223
+159,2023-04-03 00:00:00+00:00,0.002356406556685244,-0.0001489595305806355,0.0009176653826771084,-8.663952481327951e-05
+160,2023-04-04 00:00:00+00:00,-0.014246622422067666,0.0012058815462295687,0.00012426620112042258,-0.0006396890750387115
+161,2023-04-05 00:00:00+00:00,0.0006305606526503252,0.0020665270297940937,0.0014426223807864995,-0.0004797745343940435
+162,2023-04-06 00:00:00+00:00,-0.002805109031269909,0.003066165517162774,0.0016668019245809384,-0.0003731449534251712
+163,2023-04-10 00:00:00+00:00,0.013735435400105108,0.00288069576832514,0.0017993902778292862,0.0002724251663713502
+164,2023-04-11 00:00:00+00:00,0.009201902383613035,0.0020498361285647302,-3.634579558802559e-06,-0.0008399994132819134
+165,2023-04-12 00:00:00+00:00,-0.010099698021878508,-0.0009169135538235588,-0.0024300513391067036,-0.0008489977786886353
+166,2023-04-13 00:00:00+00:00,0.005428250496712833,0.0008164283218685845,-0.0017970064326388079,-0.00099840474370093
+167,2023-04-14 00:00:00+00:00,-0.003757959885105673,0.0002746810866338222,-0.0012090745656069487,-0.001104722555975088
+168,2023-04-17 00:00:00+00:00,0.009615615623868853,0.0007208890503381228,0.00035019010146842145,-0.0010270486977872206
+169,2023-04-18 00:00:00+00:00,-0.0056762236829699265,-0.002030069987767104,-0.0010858401938365063,-0.00153119996190006
+170,2023-04-19 00:00:00+00:00,-0.0014144637304471098,-0.003948818537710862,-0.001384367930451579,-0.0015277828598550722
+171,2023-04-20 00:00:00+00:00,0.002767335538123379,-0.004371467023061288,-0.002329631358592278,-0.0011637325736494279
+172,2023-04-21 00:00:00+00:00,-0.0015995437753341312,-0.0026919141540402204,-0.0023412169429187584,-0.0006985843747975502
+173,2023-04-24 00:00:00+00:00,-0.004261316252620565,-1.2132422075916005e-05,-0.0012368294266921992,-0.0009231818344172815
+174,2023-04-25 00:00:00+00:00,-0.015330205921747584,-0.0001464687475278962,-0.0016991504263254979,-0.0008547982097039953
+175,2023-04-26 00:00:00+00:00,-0.0035815642851968615,0.001150759745462075,0.0007214184566291697,-0.000578360479956867
+176,2023-04-27 00:00:00+00:00,0.011332839034553677,-0.0003335937809039507,-0.0002453970938456484,-0.0007595118405073977
+177,2023-04-28 00:00:00+00:00,0.01209668464630116,-0.002056498730682904,-0.0010425101440012525,-0.0013056366851330193
+178,2023-05-01 00:00:00+00:00,-0.004951547388086369,-0.002440221572762325,-0.0023956354466531053,-0.001547631939271932
+179,2023-05-02 00:00:00+00:00,-0.009045843080407708,-0.0032665179294349422,-0.0020051045929590948,-0.0017460450497451274
+180,2023-05-03 00:00:00+00:00,-0.011006886224849886,0.00036120829157882864,-0.0016641871980273089,-0.0013811482194951486
+181,2023-05-04 00:00:00+00:00,0.0029863923465769646,-0.00010450529153132324,6.508300172897021e-05,-0.00012800140346046917
+182,2023-05-05 00:00:00+00:00,0.010128534825498374,-5.458237791088149e-05,0.0009917209636422881,0.00046841856992471827
+183,2023-05-08 00:00:00+00:00,-0.009136219501509744,-0.00245718288322845,-0.0006404415640381346,-0.00029744731241896657
+184,2023-05-09 00:00:00+00:00,0.009134719588678399,-0.0007813572823112697,1.451784730363092e-05,0.0006676809511418545
+185,2023-05-10 00:00:00+00:00,-0.01345951252511127,-0.003724450181638795,-0.0018550824199567396,0.00047141951171005686
+186,2023-05-11 00:00:00+00:00,0.0033485663392059132,0.00023264869150747055,-0.0012372421335884874,0.0013526803986437489
+187,2023-05-12 00:00:00+00:00,-0.00208474430806864,0.0020327346613681703,-0.001573398245212833,0.0010668769695232954
+188,2023-05-15 00:00:00+00:00,-0.0005703596420971196,0.0012520961326405945,-0.0006580343205526637,0.001587392042321372
+189,2023-05-16 00:00:00+00:00,-0.00585415173007016,0.0008038751409096218,-0.0015322717742134691,0.0020830549603725137
+190,2023-05-17 00:00:00+00:00,0.0063044266332458136,4.7655153033113454e-05,-0.001149066994452006,0.0020388447043188517
+191,2023-05-18 00:00:00+00:00,0.01234842785226991,-0.002743948062611998,-0.0003918021064199368,0.0026163283368395847
+192,2023-05-19 00:00:00+00:00,-0.006053780525975995,-0.005203113338916268,-0.00011999804168715282,0.0013612889995953423
+193,2023-05-22 00:00:00+00:00,-0.0026944962988614167,-0.002693343925591055,-2.0534929884328186e-05,0.0013236447061455551
+194,2023-05-23 00:00:00+00:00,-0.009522367917327265,-0.003928326292141894,0.0012238576097094952,0.0015027709784063842
+195,2023-05-24 00:00:00+00:00,-0.007641177002339895,-0.0023421844162432493,0.002798233817540908,0.0014900550309321198
+196,2023-05-25 00:00:00+00:00,-0.0001905473120216703,0.00193197166130356,0.003915544644537916,0.0018232747112172107
+197,2023-05-26 00:00:00+00:00,0.006540152516733684,0.005059674186430057,0.0037046720419311482,0.0021907460419811954
+198,2023-05-30 00:00:00+00:00,-0.0088244545691427,0.002743089568269341,0.003778929517181903,0.002296651995244971
+199,2023-05-31 00:00:00+00:00,-0.001691616090709225,0.00652604423805196,0.005715785057655282,0.0027152150898712662
+200,2023-06-01 00:00:00+00:00,0.013868790843444482,0.008104015765929042,0.005224102488384953,0.0035898454462415376
+201,2023-06-02 00:00:00+00:00,0.015528136682794893,0.0058277250484479615,0.005606491437869229,0.0028581695227268672
+202,2023-06-05 00:00:00+00:00,-0.005016002323495917,0.0022296952884552357,0.002820090549864296,0.0021034013751807423
+203,2023-06-06 00:00:00+00:00,0.009783257567705485,0.004764002421098952,0.00264099202823996,0.0019057355354283512
+204,2023-06-07 00:00:00+00:00,0.0058390292847627055,0.0047130694133684856,0.0017654066003668317,0.0012883627151148238
+205,2023-06-08 00:00:00+00:00,0.002989780386127086,0.0023138761946370563,0.0001947415284436743,0.001192226635124865
+206,2023-06-09 00:00:00+00:00,-0.0021539266950499356,0.005272486714021355,-0.00022128981323335143,0.0016706207695048041
+207,2023-06-12 00:00:00+00:00,0.007251276348906687,0.003388512404074902,0.0006821254212582117,0.002690275907514167
+208,2023-06-13 00:00:00+00:00,0.00967122048372566,0.0005348845922808931,0.0007956158842428745,0.0021704367560073526
+209,2023-06-14 00:00:00+00:00,-0.006106718518466828,-0.0011273806604994558,-0.0002780148090930612,0.0017631684102008983
+210,2023-06-15 00:00:00+00:00,0.017630611695106983,-0.0019176576411197251,0.0018118097046250304,0.0016205063397661463
+211,2023-06-16 00:00:00+00:00,-0.011184388465247943,-0.005565190819175996,7.547616579636971e-05,0.0009053389442838483
+212,2023-06-20 00:00:00+00:00,-0.006922955190996941,-0.001990829209157894,0.001322279637752321,0.0019318286555970845
+213,2023-06-21 00:00:00+00:00,0.001277123669441989,0.0010465681947230095,0.0011166373453257404,0.0023997913041489778
+214,2023-06-22 00:00:00+00:00,-0.010096598601931683,0.0006051912730974876,0.0007800115258499162,0.00253315114438728
+215,2023-06-23 00:00:00+00:00,-0.0011157176303866906,0.005665133537497064,0.0022251863060054715,0.0030577452036504517
+216,2023-06-26 00:00:00+00:00,0.007069687543506344,0.005937109485270262,0.0035901237902501468,0.0031452220570207467
+217,2023-06-27 00:00:00+00:00,0.00828215834577142,0.004710291179203678,0.0046720430319264865,0.0027777746079635144
+218,2023-06-28 00:00:00+00:00,-0.001014645239945392,0.001208066517870794,0.003540499572992292,0.0028063883219875403
+219,2023-06-29 00:00:00+00:00,0.014687380639515161,0.0009924306610093808,0.0038555571061524596,0.002666395833051619
+220,2023-06-30 00:00:00+00:00,0.00034279922443726567,-0.0011803272038567402,0.001419575077643697,0.001820824482698578
+221,2023-07-03 00:00:00+00:00,0.0011290701215650892,0.0011641251892481896,0.0017485809139342573,0.001631352069881216
+222,2023-07-05 00:00:00+00:00,-0.008901168478966638,0.004478414058198364,0.002501496441463014,0.0011827989658368013
+223,2023-07-06 00:00:00+00:00,-0.0021395864517559766,0.0058611764627231585,0.0036565448430225887,0.001330146022227763
+224,2023-07-07 00:00:00+00:00,0.0037630209246945394,0.006732178919332426,0.004269963338990737,0.0015019982549252953
+225,2023-07-10 00:00:00+00:00,0.012190003802724817,0.004066339744761413,0.0038109255295543996,0.0011345452793088467
+226,2023-07-11 00:00:00+00:00,0.017535941836744087,0.0023210434432119494,0.002629854343745672,0.00032892300949471514
+227,2023-07-12 00:00:00+00:00,-0.002418879606997362,0.0005221616073857004,0.0008860028384278253,-0.0003438385018325311
+228,2023-07-13 00:00:00+00:00,0.0021132739831659537,0.001425529273998293,0.002016591348541809,-0.00018283128074954248
+229,2023-07-14 00:00:00+00:00,-0.008954427362971357,0.001777822478736922,0.0014283133555490618,-0.0007136023529050085
+230,2023-07-17 00:00:00+00:00,0.003547184894066345,0.003540854091934462,0.0022035983447892776,-0.0001978666447461608
+231,2023-07-18 00:00:00+00:00,0.008539705780020893,0.002965589163659028,0.001514425196322111,-0.0006362371507163049
+232,2023-07-19 00:00:00+00:00,0.0021604402672707928,0.0012868761941358736,-0.00011073104443805229,-0.0014363712203770873
+233,2023-07-20 00:00:00+00:00,0.0038461907266579116,0.002582421220273745,-0.0009590294566189845,-0.0017041006719458536
+234,2023-07-21 00:00:00+00:00,-0.00043758330710132496,0.0010854766871239015,-0.001220332336160317,-0.002705805039834434
+235,2023-07-24 00:00:00+00:00,0.0005723012112549611,0.0008546272304653712,-0.0014821189295385856,-0.0022190094763525486
+236,2023-07-25 00:00:00+00:00,0.0001743346387238252,6.567102596806448e-05,-0.0019195768362234092,-0.002260299823392191
+237,2023-07-26 00:00:00+00:00,0.008762814062311036,-0.001481212606907419,-0.0015547883493180303,-0.0024534780384044925
+238,2023-07-27 00:00:00+00:00,-0.0036649327488476146,-0.004425649673517277,-0.002346206776337394,-0.002634794257806331
+239,2023-07-28 00:00:00+00:00,-0.0015728494752666152,-0.003494842540654732,-0.002828728592795025,-0.002684004602813266
+240,2023-07-31 00:00:00+00:00,-0.003187417722142529,-0.003793575954993315,-0.002546766260857617,-0.002354303362638614
+241,2023-08-01 00:00:00+00:00,-0.007531301137287553,-0.003917048244253206,-0.0027609202356131046,-0.0020496231736035157
+242,2023-08-02 00:00:00+00:00,-0.006263170282502216,-0.001691050447785319,-0.0028147510762002066,-0.0011673403412786382
+243,2023-08-03 00:00:00+00:00,0.0010174422392713002,-0.0003193889502208177,-0.00251603211750373,-0.0007131875761767107
+244,2023-08-04 00:00:00+00:00,-0.0032251198312741296,-0.0022572249259871684,-0.004272105528371576,-0.0007080062535554283
+245,2023-08-07 00:00:00+00:00,-0.003902053400111325,-0.001377039587405036,-0.0030074041394710406,-0.000821988694787991
+246,2023-08-08 00:00:00+00:00,0.003899624840630696,-0.0016465021421528422,-0.002643138096912309,-0.0012188914085903531
+247,2023-08-09 00:00:00+00:00,0.0006596539338907481,-0.003985762949633756,-0.0033696723191994423,-0.001602560549650205
+248,2023-08-10 00:00:00+00:00,-0.008688326473253582,-0.004752326285689679,-0.002983565005413988,-0.0016085353313492518
+249,2023-08-11 00:00:00+00:00,0.0011245334529001738,-0.006365486911837098,-0.0026109195927801585,-0.0009313562953467805
+250,2023-08-14 00:00:00+00:00,-0.0051615847773311225,-0.004671905190308137,-0.002199311429380531,-0.001229843661781022
+251,2023-08-15 00:00:00+00:00,-0.007981375270576718,-0.003684029521495181,-0.0013751842778620782,-0.000879653751249868
+252,2023-08-16 00:00:00+00:00,-0.0032064839662822027,-0.002828684678701298,0.0005252993769284138,-0.0003553164666834776
+253,2023-08-17 00:00:00+00:00,-0.01685131624010656,-0.001254677006237483,0.001142192467334756,-0.00014789496647458832
+254,2023-08-18 00:00:00+00:00,0.009945849601211155,0.0011738633893805808,0.003012279841578663,0.00045847221538058964
+255,2023-08-21 00:00:00+00:00,-0.00018904203294339795,0.0002742703947120698,0.0014139213300489875,-0.00018808798573131719
+256,2023-08-22 00:00:00+00:00,-0.003666976380894478,0.0009558735025072285,0.00019111841931518148,-4.991336872233928e-05
+257,2023-08-23 00:00:00+00:00,0.004798775042443972,0.003959739863446514,0.00017541459452963726,-0.0005074693272879067
+258,2023-08-24 00:00:00+00:00,-0.004701932743806158,0.003572163901035064,-0.0002626939208185824,-0.001266906673814846
+259,2023-08-25 00:00:00+00:00,0.0053660924878159615,0.004796205995451227,0.0007523310745652684,-0.0015671915499834738
+260,2023-08-28 00:00:00+00:00,0.0032452563566078754,0.002548793625072,-0.00028397791027260006,-0.001837479232745608
+261,2023-08-29 00:00:00+00:00,0.011122534949396442,-0.0005664398766606294,-0.0004017946934881293,-0.0021977040066625585
+262,2023-08-30 00:00:00+00:00,0.002885672237027551,-0.0035505984206576926,-0.0012373589875797147,-0.00288995724310355
+263,2023-08-31 00:00:00+00:00,0.0011478537106194588,-0.004025574940998672,-0.0014172105919824749,-0.00242013555253556
+264,2023-09-01 00:00:00+00:00,-0.005726262443671057,-0.0032134721458900667,-0.0019974581448265745,-0.0030658330266932136
+265,2023-09-05 00:00:00+00:00,-0.01218722088523481,-0.0030277974545576637,-0.0017287420965494793,-0.003533075249154455
+266,2023-09-06 00:00:00+00:00,-0.003906772008041131,-0.00021230142154292117,-0.00025745275593434873,-0.003140222226237135
+267,2023-09-07 00:00:00+00:00,0.0004020886904597348,0.0010900110800868674,-0.0011616212353748032,-0.002920520479548895
+268,2023-09-08 00:00:00+00:00,0.005305309867926114,0.0012343005384163183,-0.0022606398369324925,-0.0033164528697848935

factor_data.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:904389c06f93b20202bcd0d43f2a9caaa39d5efdf3c9789da904cb71c634ce96
+size 11926919

financial_analysis.py

@@ -0,0 +1,205 @@
+import pandas as pd
+import numpy as np
+import yfinance as yf
+import ast
+from scipy import stats
+from datetime import datetime, timedelta
+import pytz
+import pandas_market_calendars as mcal
+import alphalens as al
+import matplotlib.pyplot as plt
+
+def sentiment_to_numerical(sentiment):
+    mapping = {'Negative': -1, 'Positive': 1, 'Neutral': 0}
+    return sentiment.map(mapping)
+            
+def process_sentiment_data(sentiment_data = 'finbert_sentiment.csv', sector_ticker = 'sector_ticker.csv', prices = 'prices.csv'):
+    columns_to_load = ['Ticker', 'pubDate', 'finbert_output']
+    df = pd.read_csv(sentiment_data, usecols=columns_to_load)
+    df.rename(columns={'Publication Date': 'pubDate','finbert_output': 'Sentiment'}, inplace=True)
+
+    # Adjusting the dates of news articles
+    nyse = mcal.get_calendar('NYSE')
+
+    trading_start_hour = 9
+    trading_start_minute = 30
+    trading_end_hour = 16
+    trading_end_minute = 0
+
+    def adjust_date(pub_date):
+        if pd.isnull(pub_date) or not isinstance(pub_date, pd.Timestamp):
+            return pub_date
+
+        trading_start_time = pd.Timestamp(f'{pub_date.date()} {trading_start_hour}:{trading_start_minute}')
+        if pub_date >= trading_start_time:
+            next_trading_day = nyse.schedule(start_date=pub_date.date() + pd.DateOffset(days=1), end_date=pub_date.date() + pd.DateOffset(days=10)).iloc[0]['market_open']
+            return next_trading_day
+        else:
+            valid_days = nyse.valid_days(start_date=pub_date.date(), end_date=pub_date.date())
+            if not valid_days.empty and pub_date.date() == valid_days[0].date():
+                return pub_date
+            else:
+                next_trading_day = nyse.schedule(start_date=pub_date.date() + pd.DateOffset(days=1), end_date=pub_date.date() + pd.DateOffset(days=10)).iloc[0]['market_open']
+                return next_trading_day
+
+    df['pubDate'] = df['pubDate'].apply(adjust_date)
+
+    # Converting probabiltiies to one value
+    def convert_sentiment_to_score(sentiment):
+        predicted_sentiment_probabilities = {}
+
+        components = sentiment.split(', ')
+        for component in components:
+            key_value = component.split(':')
+            if len(key_value) == 2:  
+                key, value = key_value
+                key = key.strip(" '{}").capitalize()  
+                try:
+                    value = float(value.strip())  
+                except ValueError:
+                    continue  
+                predicted_sentiment_probabilities[key] = value
+        
+        positive = predicted_sentiment_probabilities.get('Positive', 0)
+        negative = predicted_sentiment_probabilities.get('Negative', 0)
+        neutral = predicted_sentiment_probabilities.get('Neutral',0)
+        sentiment_score = (positive - negative)/(1 + neutral)
+        return sentiment_score
+
+    df['Sentiment_Score_2'] = df['Sentiment'].apply(convert_sentiment_to_score)
+
+    # replacing invalid tickers
+    df['pubDate'] = pd.to_datetime(df['pubDate'], utc=True, format='ISO8601')
+    df['pubDate'] = df['pubDate'].dt.date
+
+    print(df['pubDate'].dtypes)
+
+    replacements = {
+        'ATVI': 'ATVIX',
+        'ABC': 'ABG',
+        'FBHS': 'FBIN',
+        'FISV': 'FI',
+        'FRC': 'FRCB',
+        'NLOK': 'SYM.MU',
+        'PKI': 'PKN.SG',
+        'RE': 'EG',
+        'SIVB': 'SIVBQ',
+    }
+
+    df['Ticker'] = df['Ticker'].replace(replacements)
+
+    df = df[df['Ticker'] != 'SBNY'] 
+
+    #
+    aggregated_data = df.groupby(['Ticker', 'pubDate'])['Sentiment_Score_2'].mean().reset_index()
+    aggregated_data['pubDate'] = pd.to_datetime(aggregated_data['pubDate']).dt.tz_localize('UTC')
+    aggregated_data.set_index(['pubDate', 'Ticker'], inplace=True)
+
+    prices = pd.read_csv(prices, index_col=0, parse_dates=True)
+
+    #
+    equal_weighted_benchmark = prices.pct_change(periods=1).shift(periods=-1).mean(axis=1)
+
+    equal_weighted_benchmark_df = equal_weighted_benchmark.reset_index()
+    equal_weighted_benchmark_df.columns = ['date', 'equal_weighted_benchmark']
+    returns_5d=prices.pct_change(periods=5).shift(periods=-5)/5
+    returns_10d=prices.pct_change(periods=10).shift(periods=-10)/10
+    returns_20d=prices.pct_change(periods=20).shift(periods=-20)/20
+    mean_5d = returns_5d.mean(axis=1).reset_index()
+    mean_10d = returns_10d.mean(axis=1).reset_index()
+    mean_20d = returns_20d.mean(axis=1).reset_index()
+    mean_5d.columns = ['date', '5d_mean_return']
+    mean_10d.columns = ['date', '10d_mean_return']
+    mean_20d.columns = ['date', '20d_mean_return']
+    equal_weighted_benchmark_df = equal_weighted_benchmark_df.merge(mean_5d, on='date', how='left')
+    equal_weighted_benchmark_df = equal_weighted_benchmark_df.merge(mean_10d, on='date', how='left')
+    equal_weighted_benchmark_df = equal_weighted_benchmark_df.merge(mean_20d, on='date', how='left')
+    cut_date_min= aggregated_data.index.get_level_values('pubDate').min()
+    cut_date_max= aggregated_data.index.get_level_values('pubDate').max()
+    equal_weighted_benchmark_df = equal_weighted_benchmark_df[equal_weighted_benchmark_df.date>=cut_date_min]
+    equal_weighted_benchmark_df = equal_weighted_benchmark_df[equal_weighted_benchmark_df.date<=cut_date_max]
+    equal_weighted_benchmark_df
+
+    #
+    tickers = aggregated_data.index.get_level_values('Ticker').unique()
+    start_date = aggregated_data.index.get_level_values('pubDate').min() - pd.Timedelta(days=30)
+    end_date = aggregated_data.index.get_level_values('pubDate').max() + pd.Timedelta(days=30)
+
+    all_dates = prices.loc[cut_date_min:cut_date_max].index
+    all_tickers_dates = pd.MultiIndex.from_product([tickers, all_dates], names=['Ticker', 'Date'])
+    all_tickers_dates_df = pd.DataFrame(index=all_tickers_dates).reset_index()
+    aggregated_data_reset = aggregated_data.reset_index()
+    merged_data = pd.merge(all_tickers_dates_df, aggregated_data_reset, how='left', left_on=['Ticker', 'Date'], right_on=['Ticker', 'pubDate'])
+    sector_data = pd.read_excel('scraping.xlsx', usecols=['Ticker', 'Sector'])
+    merged_data = merged_data.reset_index()
+    merged_data = pd.merge(merged_data, sector_data, how='left', left_on='Ticker', right_on='Ticker')
+
+    #
+    decay_factor = 0.7
+
+    for ticker in tickers:
+        ticker_data = merged_data[merged_data['Ticker'] == ticker].copy()
+
+        original_nans = ticker_data['Sentiment_Score_2'].isna()
+
+        ticker_data['Sentiment_Score_2'] = ticker_data['Sentiment_Score_2'].ffill()
+
+        for i in range(1, len(ticker_data)):
+            if original_nans.iloc[i]:
+                ticker_data.iloc[i, ticker_data.columns.get_loc('Sentiment_Score_2')] = ticker_data.iloc[i - 1, ticker_data.columns.get_loc('Sentiment_Score_2')] * decay_factor
+
+        merged_data.loc[merged_data['Ticker'] == ticker, 'Sentiment_Score_2'] = ticker_data['Sentiment_Score_2']
+
+    merged_data['Sentiment_Score_2'].fillna(0, inplace=True)
+    merged_data.drop(columns=['pubDate'], inplace=True)
+    merged_data.set_index(['Date', 'Ticker'], inplace=True)
+
+    return merged_data, prices, equal_weighted_benchmark_df
+
+# Alphalens
+def alphalens_analysis(merged_data, prices):
+    factor_data=[]
+    factor_data = al.utils.get_clean_factor_and_forward_returns(
+        factor=merged_data['Sentiment_Score_2'],
+        prices=prices,
+        binning_by_group=False,
+        bins=None,
+        quantiles=5,
+        periods=(1, 5, 10, 20),
+        groupby=merged_data['Sector'],
+    )
+
+    al.tears.create_returns_tear_sheet(factor_data, long_short=True, group_neutral=False)
+
+    return factor_data
+
+def alphalens_analysis_by_sector(factor_data):
+    mean_return_by_qt, std_err_by_qt = al.performance.mean_return_by_quantile(factor_data, by_group=True)
+    al.plotting.plot_quantile_returns_bar(mean_return_by_qt, by_group=True)
+
+def calculate_information_ratio(factor_data, equal_weighted_benchmark_df):
+    # Merge the factor data with the benchmark data
+    factor_data = factor_data.merge(equal_weighted_benchmark_df, on='date', how='left')
+    
+    # Calculate excess returns for various holding periods
+    factor_data['excess_return_1D'] = factor_data['1D'] - factor_data['equal_weighted_benchmark']
+    factor_data['excess_return_5D'] = factor_data['5D'] - factor_data['5d_mean_return']
+    factor_data['excess_return_10D'] = factor_data['10D'] - factor_data['10d_mean_return']
+    factor_data['excess_return_20D'] = factor_data['20D'] - factor_data['20d_mean_return']
+    
+    # Initialize a DataFrame to store IR results
+    results = pd.DataFrame(index=range(1, 6), columns=['IR 1D', 'IR 5D', 'IR 10D', 'IR 20D'])
+    
+    # Calculate IR for each quantile and holding period
+    for quantile in range(1, 6):
+        for period in [1, 5, 10, 20]:
+            column_name = f'excess_return_{period}D'
+            tmp = factor_data[factor_data.factor_quantile == quantile][['date', column_name]].groupby('date').mean()
+            ir = np.mean(tmp) / np.std(tmp) * np.sqrt(252)
+            results.at[quantile, f'IR {period}D'] = ir.values[0]
+    
+    from IPython.display import display
+    display(results.style.format("{:.3f}"))
+    
+
+

finbert_sentiment.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:149783b38a6b94ad2409cbe50dcac62e915dbb7cc0143b3154dc8199d4463423
+size 98368182

requirements.txt

@@ -3,4 +3,8 @@ pandas
 numpy
 openai
 tqdm
-matplotlib
+matplotlib
+scipy
+ast
+yfinance
+alphalens