Skip to content

dspytdao/Heroku_Graph_ML

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

63 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Heroku_Graph_ML

Deployed Using Heroku at https://dspyt.herokuapp.com/

We are accessing Uniswap V3 GraphQL at https://thegraph.com/hosted-service/subgraph/uniswap/uniswap-v3?selected=playground to obtain historical data for the chosen token. The Default token is Uniswap.

Next we model the time-series data to predict token price for the N number of timestamps. Default is 15 periods.

Model

ARIMA General Case

We have Used Arima(1,1,0) model since it has maths foundation and differences the data to make it stationary.

In particular properties of AR(1) are as following: image

We are also using the app to create a larger app for the Uniswap bounty. The app output image

Further References

DSPYT: Time series data – An easy introduction

DSPYT: Simple Return, Log Return and Volatility easy intro

DSPYT: Advanced Realized Volatility and Quarticity

DSPYT: Machine Learning simple Sklearn Ensemble