Giter VIP home page Giter VIP logo

devalpha-node's Introduction

DevAlpha

Build Status Dependencies NPM Version Coverage Status

DevAlpha is a Javascript framework for creating and running your own algorithmic trading systems. It is built using TypeScript, weighs in at a less than 1500 lines of code, and is speedy as hell.

The internal architecture primarily consists of one big stream and a bunch of consumers. It is implemented using the excellent Highland streams library, and also makes use of some helper functions from Redux.

Features

  • Event sourced
  • Tiny footprint
  • Easily extensible
  • Simple API
  • Thoroughly tested
  • Typescript definitions

Installation

Install using NPM:

npm install devalpha

Quickstart

(Check out devalpha-example for a starter repo.)

Getting started is easy as pie. Hook up any source of data you like and start trading in seconds.

import { createTrader } from 'devalpha'

const feeds = {
  myQuandlFeed: [1, 2, 3, 4],
  myStreamFeed: fs.createReadStream(...)
}

const strategy = (context, action) => {

  // Place an order
  if (action.type === 'myQuandlFeed') {
    context.order({
      identifier: action.payload.identifier,
      quantity: 100 * action.payload.signalStrength,
      price: 1000
    })
  }

  // Get current portfolio state
  const state = context.state()

  // Cancel an order
  if (state.capital.cash < 10000) {
    context.cancel({
      id: 123
    })
  }
}

// Create the trading stream
const stream = createTrader({ feeds }, strategy)

// Consumer the stream and make money!
stream.done(() => {
  console.log('Finished!')
})

Settings

const settings = {

  /* Toggle backtesting/realtime mode. In backtesting mode, events from the feed stream are pulled
  as needed rather than pushed as created. This allows you to do a number of events for each feed
  item, and then pull the next one only when you're finished with the current.
  
  NOTE: DevAlpha will only activate realtime mode when this parameter is explicitly set to `false`.
  This means that setting `backtesting: 0` will not do the job. */
  backtesting: true,

  /* Only used in realtime mode. The client manages order execution, and is provided to the 
  internal broker middleware. */
  client: null,

  /* Define the starting capital of your algorithm. Use only in backtesting mode. In realtime mode
  you're better of using the `initialStates` setting instead. */
  startCapital: 0,
  
  /* Provide initial states for your algorithm. One obvious use case would be when realtime
  trading, and you want to fetch positions, capital, or order information from your broker. */
  initialStates: {},

  /* An object mapping event names to stream-like objects. See https://highlandjs.org/#_(source)
  for a definition of "stream-like". Keys will be used as event type names. */
  feeds: {},

  /* Settings for your backtest. */
  backtest: {
    
    /* Denotes when your backtest is started (the first date of your backtesting data). */
    timestamp: 0,

    /* A number or a function used when calculating expected commission. */
    commission: 0

  },

  /* Settings for the guard middleware, which will prevent or alter orders (based on your
  configuration). */
  guard: {
    
    /* Allow/disallow shorting. */
    shorting: false,
    
    /* Allow/disallow trading on margin. */
    margin: false,

    /* An array of restricted instrument identifiers. Example: ['GOOG', 'SPOT']. */
    restricted: []

  },

  /* DevAlpha dashboard settings. */
  dashboard: {

    /* Toggle the DevAlpha dashboard. */
    active: false,

    /* Port used to pipe portfolio data. */
    port: 4449
  }

}

Usage

The createTrader-function returns an unconsumed stream, and so it is up to you to consume it (thereby running the strategy). Highland provides a number of ways of doing this (see here), but the easiest one is probably just to use .resume() like so:

const settings = {...}
const strategy = (context, action) => {...}

createTrader(settings, strategy).resume()

However, you could also do crazy things like this:

import { createTrader, ORDER_FILLED, ORDER_FAILED } from 'devalpha'

const settings = {...}
const strategy = (context, action) => {...}

const stream = createTrader(settings, strategy)

const slackStream = stream.fork()
const redisStream = stream.fork()

// Get Slack notifications on filled or failed orders
slackStream.each((event) => {
  if (event.action.type === ORDER_FILLED) {
    slackLogger.log('Hooray! An order was filled!')
  } else if (event.action.type === ORDER_FAILED) {
    slackLogger.log('Whoops! One of your orders was not executed!')
  }
})

// Place the current state in a Redis store
redisStream.each((event) => {
  redisClient.set('state', JSON.stringify(event.state))
})

Pretty neat, huh?

Resources

License

GNU GPL license. See the LICENSE.md file for details.

Responsibilities

The author of this software is not responsible for any indirect damages (foreseeable or unforeseeable), such as, if necessary, loss or alteration of or fraudulent access to data, accidental transmission of viruses or of any other harmful element, loss of profits or opportunities, the cost of replacement goods and services or the attitude and behavior of a third party.

devalpha-node's People

Contributors

fhqvst avatar

Watchers

James Cloos avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.