Team Quantino believe in a world where everyday users have the power to forecast financial markets at the tip of their fingers, by democratising the data and forecasting techniques previously available only to experts.
The PoC solution augments traditional forecasting techniques with RNN (Recurrent Neural Networking) deep learning algorithms and infinitely scalable serverless compute. Tech stack consists of AWS Lambda, AWS S3, Anodot on AWS, React Native/Android application.
Team Quantino consists of Yun Zhi Lin, Head of Engineering (Contino), Lucas Rafagnin, Cloud Lead (Contino), Ira Cohen, Chief Data Scientist (Anodot), Sami "The Machine" Raines, Data Engineer (Contino), Raymond Au, Data Engineer (Contino)
Financial Forecasting using Recurrent Neural Network, Social Media and Cloud
Using RNN, Social Media and Cloud
Yun Zhi Lin
Head of Engineering
Team Quantino Global
Evolution of (Financial) Forecasting
Oracle Bones Box and Jenkins
Excel Machine Learning
Social Media and Trump Driven Data
$1.3 billion wipe out overnight
But Your Users are not Data Scientists ...
But Your Users deserve Machine Learning
Training and Algorithms Under the Hood
Model: Hybrid RNN & Holt-Winters
● Rolling 14 day forecasts based on historical daily High, Low, Adjusted Closing Price and Volume
● Symmetric Mean Absolute Percent Error (SMAPE) of < 0.01 % for big four banks, from Dec 29th
2017 to Nov 21st 2018
● Presented at NAB 700 on Nov 7th 2018 as part of Data Showdown meetup event