Macro Events impact the Business function of CITI because of its geopolitical volatility and unpredictability. When Macro events trigger the market and have a long-term impact it is considered as a โ Macro Trend โ which has a ripple effect on Financial Markets. Oftentimes, the Potential Impact is realized too late.!
http://ec2-23-22-153-152.compute-1.amazonaws.com/
To develop the model which predicts the factors affecting the market movement events ahead of the News Channels across the world. Grouping of Events in a Trend which will affect the market.
-
We have created a web application that collects real-time data from various sources like Twitter, news API, etc., and provides insights on which asset classes will impact the market in the near future.
-
Our approach is to identify the micro trend happening in social networking platforms and then analyzing with our asset classes ( metadata) to figure out the macro trend.
-
Our market prediction is based on 9 asset classes which will have a potential impact on the financial market across the globe.
- This feature is password protected as there are limitations of free twitter and news api developer account.
- Once a valid password is provided, the app polls for livestream of feeds from twitter and news api.
- The current configuration makes 1 request every 4 seconds which pulls around 250 news feeds from the internet.
- After 6 requests the live stream is forcefully stopped (as twitter and news api may block the account for extracting a large chunk of data within a short span of time).
- Everytime we do a live stream we generate processed and cleaned files.
- These cleaned files are used to create the trends whenever user clicks on previous trends option.
- Clone the code on your local system using
git clone https://github.com/uanthwal/citi-hackathon.git
- Navigate to FrontEnd folder
- Install all the dependencies using
npm install
- Start the angular app using
ng serve
- As the code is already clone in previous steps, navigate to Backend folder
- Install all the dependencies using
pip3 install -r requirements.txt
- Run the python local server using command
python3 data_extractor.py
- App server will be available on
http://localhost:5000/
- NOTE: The twitter and news APIs keys have been removed from the code. In order to make the backend server running the keys are required.
-
The Asset Classes metadata can further be extended so that all the High-Frequency Identifiers can easily fit into any particular Asset class or all the Asset Classes.
-
The sentiments behind the tweets or posts on any social platform can be further analyzed to identify its impact on asset classes more accurately.
-
The data can be extracted from various social platforms like LinkedIn etc., and integrated to identify microtrends from bigger and diverse population.
- https://twitter.com/
- https://github.com/
- analyze word frequency counts using twitter data and tweepy in python | earth data science - earth lab
- https://newsapi.org/
- https://fetchrss.com/api
- https://developer.nytimes.com/
Got any questions for us? Please drop a mail to one of us, and we would be more than happy to revert. You can find us at [email protected] , [email protected] , [email protected]