An EFT powered by NLP and Sentiment Analysis

This project is aimed at utilizing NLP, ML and cutting-edge sentiment analysis techniques to analyse data from Twitter API, and News API to inform investment decision in order to build and backtest an ETF.

DEEGA-S ETF (the “Fund”) is intended to be an actively managed blend ETF that seeks to replicate DEGA’s proprietary US equity sentiment Index with an aim to beat S&P 500 (the “Benchmark) before fees and expenses.

Under normal conditions the Fund intends to invest 100% of its net assets in an equally weighted in securities that comprise top 10 of the DEGA’s US Equity Sentiment Index constructed using a proprietary methodology. The companies that qualify to be included in this index are all companies traded on three principal indexes (S&P 500, DJIA, and NASDAQ) with a market cap > USD 2 Billion. Since this is a sentiment bases active investment strategy, the fund managers review the portfolio on a regular basis and rebalance the holdings to reflect the latest composition of the Equity Sentiment Index.

David Costa, Evita Louissaint, Esteban Cervantes, Gabriela Galarza, Abhishek Srivastava.

The portfolio performance is based on a 14-day trading cycle. The fund has managed to meet all its targets set by the portfolio managers, including beating the benchmark S&P 500.

a) Concentration Risk: Since market sentiments tend to, at times, favor certain sectors or economy, there is chance that the Funds’ investment can get crowded into a particular sector. The regular reviews of the sector will help in mitigating this risk. b) Market Risk: The investments are exposed subject to the market risk like all investments. The diverse nature of the portfolio will help in mitigating a few of the risks. c) New Fund Risk: The Fund is new type of fund with limited or no operational history and a small asset base. d) Risk of social media analytics: Social media analytics aim

This ETF is for only an academic exercise and should not be taken as an investment advice.

first Part
pip install tweepy
pip install yfinance 
pip install -U python-dotenv
pip install -U textblob
pip install seaborn
pip install matplotlib.pyplot
pip install regex
pip install requests
pip install datetime
pip install alpaca-trade-api
pip install pandas
News API
	Vader Model
	BERT model
Second Part

• DEGA tool- a tool to save tweets
• Notebook to News Sentiment
• Notebook to News_BERT
• Notebook to News RNN
• Notebook Ticker Grabber Tool
• Notebook Buy sell
• Notebook Portfolio Returns
• Notebook Portfolio Returns with dashboard
• Notebook to Twitter Sentiment
• Notebook to Twitter NLP

This ETF is build upon a proof of concept exercise conducted under:

• Project 1