QuantResearchDev by Moon Hedge Fund

Wanna talk with us ? : https://discord.gg/Qq4MVVV

Quant Research dev & Traders open source project

Who is Moon ?

A community of traders & coders that want to spread some good logic to improve the overral knowledge of automated trading (no identity nor invite will ever be provided to join us, we are closed)

User stories

• I want to be able to run backtests against 1 coin pair
• I want to be abke to run backtests against multiple coins
• I want to be able to run backtrsts using different candle sizes
• I want to be able to implement multiple indicators in a single strategy using multiple time frames
• I want to be able to see the order book in the strategy
• As a user who wants to create some strat, I want to be able to easily draw some graphs with some custom info ( indicators / max / min / Sell / buy / anything )(edited)
• As a user I want this bot to run on several coins
• As a user I want fine tuned buy and sell signal, at sub minute timings, ( probably though web sockets )
• as a user, I want an integration with somethign like discord or telegram, so I can get some information from my running bot from my phone
• I would like to be able to monitor account holdings
• I would like to be able to manage the asks and bids volume based on my own rules
• i want: function of strategy responded for currency/fiat . ex: btc in last 2 minutes -2% then sell asset, currency to USDT
• i want: strategy for execution orders (can write do use market, limit and when - high volume, retries failed etc)
• As a strat explorer, I want to have several strat running on paper trade at the same time

Concept

• Rebuilt time candle dynamicaly to expand view
• Invoke dynamicaly strategy by weight score
• Use direct websocket & Azure Data Table stored data

Data Model

Concept

• IDWN
• IExchanger -> Platform
• IIndicator -> TA virtual method override
• IIMarket -> Stream all market pairs to built sentiment, retrace and corelate news
• INode -> Restream to node using websocket all moon roles
• IInterface -> Use modelized view for UI rendering
• Performance -> manage resources allocation and repeated task through timespan
• Risk -> Risk model using RMSI (Tree based risk management) (Identify - Avoid - Hold - Transfer)
• Risk Parking -> Parking hodler statisticals follower and manager
• DNA -> Dynamic code agent to change runtime logic
• RNN -> Recurent Neural Network to predict slope / direction

Interface Overview

Real Time wall bounce overview

Market Automatic Pairs detections

Real time market broadcaster

Real time trady trading pattern & ta in server node

Real time binance market all pairs movement

Real Time overall market sentiment and capital

Azure Data Table storage

Auto elect pair for a given strategy before live

Some custom TA

Release v0.1

Best effort for EOY 2018.

Credits

• Bitmex (for the good market making codes)
• Bitfinex (for the code websocket srv)
• Trady (with love)
• Ninja (for the good design)