Skip to content

Learning Resources

Source: Notion | Last edited: 2023-09-26 | ID: 703a6fe2-822...


⭐ The books we recommend are a gateway—a key that will unlock a wealth of knowledge and understanding. And to gain access to these tomes, you must seek out the guardian of this treasure trove—⁣Contact DO. Only he can grant you entrance to the hallowed halls of the Guest Researcher Area workplace on Notion. And within those walls, you will find not only the books you seek, but also the holy grail of data—Binance historical sampled LOB on Amazon S3.

Through careful study and dedication, you will be able to navigate the complex landscape of automated financial markets with greater confidence and skill. Do not be discouraged if you encounter obstacles along the way, for with perseverance and determination, you will surely master this powerful tool. Whether you are an experienced trader or a newcomer seeking to expand your knowledge, these articles offer valuable insights that you can use to your advantage in the volatile world of finance. So arm yourself with a thirst for knowledge and embark upon this enlightening journey. With a bit of luck and a lot of hard work, you will emerge as a more successful and savvy trader, ready to take on the markets and emerge victorious.

[https://github.com/jo-cho/trading-rules-using-machine-learning](https://github.com/jo-cho/trading-rules-using-machine-learning) generate trading strategy by using [https://github.com/autogluon/autogluon](https://github.com/autogluon/autogluon)

For time-series data containing multiple individual series, AutoGluon can produce forecasting models to predict future values of each series based on historical observations of both this series and all of the other series in the dataset. A single call to AutoGluon TimeSeriesPredictor’s fit() automatically trains multiple models on a dataset containing multiple time-series measured over the same time period, and does not require you to manually deal with cumbersome issues like data cleaning, hyperparameter optimization, model selection, etc. Most neural network-based models are from the GluonTS library. Allowed to contain missing values and additional (non-time-varying) static features, the data can be loaded from: a CSV file or the GluonTS format.

https://auto.gluon.ai/dev/tutorials/timeseries/index.html

[https://github.com/KodAgge/Reinforcement-Learning-for-Market-Making](https://github.com/KodAgge/Reinforcement-Learning-for-Market-Making)- RL+MM project
[https://github.com/roq-trading/roq-api](https://github.com/roq-trading/roq-api) - [https://github.com/roq-trading/roq-samples](https://github.com/roq-trading/roq-samples) - [https://github.com/roq-trading/roq-python](https://github.com/roq-trading/roq-python) - A toolkit for quant traders wanting full control of their own trading platform.

Open, modular and built for ultra-low latency market making. An ambitious platform that comes with Back-Testing, Historical Simulation, and consultation services. Here’s their documentation site

https://roq-trading.com/docs/

[https://github.com/hummingbot/hummingbot](https://github.com/hummingbot/hummingbot) build high-frequency crypto trading bots that specialize in market making and arbitrage strategies using Avellaneda-Stoikov market-making algorithm

https://blog.hummingbot.org/2022-03-02-beginners-top-misconceptions/

Video

♐ Other open source implementation of Avellaneda-Stoikov market-making algorithm:

https://medium.com/open-crypto-market-data-initiative/simplified-avellaneda-stoikov-market-making-608b9d437403

♐ Other open source implementation of Avellaneda-Stoikov market-making algorithm:
[https://github.com/dcajasn/Riskfolio-Lib](https://github.com/dcajasn/Riskfolio-Lib) can be useful if you have multiple strategies to allocate. Portfolio Optimization and Quantitative Strategic Asset Allocation in Python
[https://github.com/ccxt/ccxt](https://github.com/ccxt/ccxt) is the most popular one-stop cryptocurrency trading API. But it’s unlikely to be useful unless your strategy trade on multiple exchanges at the same time
[https://github.com/freqtrade/freqtrade](https://github.com/freqtrade/freqtrade)ambitious crypto trading bot project, good for charting features and examining executions.

https://www.ml-quant.com/

https://quantocracy.com/

  • 2023-01-12 Suggested by Thai Nguyen: Causal Factor Investing: Can Factor Investing Become Scientific? - López de Prado version: Jan 8, 2023
2022-12-24 Found Avellaneda-Stoikov market-making algorithm through this December 20, 2022 paper. It seems to be the de facto standard for market making:

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0277042

  • Cascaded LSTM Networks, J Zou, J Lou, B Wang, S Liu, 2022

  • Why machine learning “succeeds” in development but fails in deployment — causaLens

  • Causality for Machine Learning FF13 · May 2020 — Cloudera Fast Forward Labs Research

  • Artificial Intelligence and Causal Inference — 2022, Momiao Xiong

  • https://akunacapital.com/ - Akuna Capital is a proprietary trading firm specializing in derivatives market making and sophisticated modeling with a commitment to cutting-edge technology.
Kbit — Trading Algo Developer (New York) 2022-11-21 [https://www.linkedin.com/company/kbitllc/about/](https://www.linkedin.com/company/kbitllc/about/)
  • Founder and CEO We perform market making and high frequency trading of cryptocurrencies, trading tens or hundreds of millions of dollars worth of cryptos daily. We trade cryptos 24/7 on a globally distributed computational platform. We’re applying software engineering, data engineering, and financial engineering to the crypto markets. Using our proprietary technology platform we perform research, opportunity identification, market and counterparty risk management, and trade execution. Our strategies include market making, directional forecasting, and arbitrage. We take a scientific approach in identifying and capturing trading opportunities. We hypothesize, research, confirm, implement. And we’ve proven the ability to make profits in up and down markets.
  • Marc Zeitouni — Partner and CTO/COO at Kbit — https://www.linkedin.com/in/marc-zeitouni-36973861/ — We are a leader in cryptocurrency high frequency trading and trade roughly 1% of global cryptocurrency volume every day, managing $75m+ in AUM. We trade cryptos on global exchanges 24/7 via a globally distributed computational platform. Our strategies include market making, directional forecasting, and arbitrage.

https://www.indeed.com/jobs?q=algo+trading&start=10&pp=gQAPAAAAAAAAAAAAAAAB7okFOgAVAQAA2LczJQk9s74s78oRSfCXGRUGAAA&vjk=f267957e2c929d63

  • We are looking for engineers of all levels to help us enhance and expand the company’s trading platform. The successful candidate will help maintain and enhance our existing platform, while simultaneously helping us grow and expand our platform to facilitate the company’s growth and vision to be a leading high-frequency firm in crypto. We have basically no UIs, all development is back-end relating to our automated trading system. This includes things like algorithmic trading, connections to exchanges, data collection and processing, and model analysis and experimentation.
  • Bachelor’s degree in Computer Science, Mathematics, or related field
  • Understanding of computer science data structures
  • Experience programming in an object-oriented language
  • Intellectually curious
  • Self-motivated and driven
  • Live in the New York, NY area- This position is on-site
  • Experience working on trading systems
  • Understanding of data structures
  • Distributed software architecture
  • Languages: Python, Java, Go, C++
  • Platforms: Linux, AWS Kbit is a quantitative, high-frequency, high-turnover cryptocurrency trading firm. Founded almost 5 years ago, we perform hundreds of thousands of trades a day in cryptos and crypto derivatives globally using a proprietary, distributed technology platform. Since inception we’ve done over 70billionintradingvolumeandcurrentlytradearound70 billion in trading volume and currently trade around 5 billion a month across many exchanges. We’ve been consistently profitable over different market cycles during that time. We’ve built a proprietary platform for market data processing, order management, strategy execution, research, accounting, and operations.

We are a very small and entrepreneurial team. Our culture is casual, scientific, and engineering focused. There’s zero red tape, and tons of opportunity for impact. Our capital base is growing materially, and we are looking to expand our trading strategies and capabilities.

  • **Exegy Xero VTE — **https://www.youtube.com/watch?v=sOm8q1UkeuU provides tick-to-trade latency performance is under 120 nanoseconds for canceling and under 250 nanoseconds for mass quoting and aggressing orders.
  • BSOhttps://www.bso.co/crypto-trading-solutions?hsLang=en provide low latency, cloud connections from data centres to cloud and between clouds connecting crypto firms to exchanges while also providing the necessary global reach to onboard new exchanges as soon as they come online.

http://trality.com allows python-like scripting. They offers existing template for multiple crypto instruments as signal inputs so that we can backtest the validity of divergence strategy of correlated instruments. 2021-10-15

https://www.quantreex.com/ offers Scratch-like colored block programming. But unfit for most of our purposes. Ignore. 2021-10-15

https://contentapi.tradermade.com/documentation TradeMade offers API but has only a few crypto pairs. Asked their customer services when they’ll have more crypto instrument available. 2021-10-16

https://www.letsdeel.com/

2023-02-25 FTX’s downfall and Binance’s consolidation: the fragility of Centralized Digital Finance, by David Vidal-Tomás, Antonio Briola, Tomaso Aste https://arxiv.org/abs/2302.11371

2022-11-21 — Proof of Reserves** — Coinbase Pro, Bitfinex, Gemini, and Kraken are the next fourth-largest exchange holders of Bitcoin after Binance, as per Coinglass, respectively holding 528,900, 345,597, 153,212, and 70,622 coins. Among the top five exchanges, Binance is the only one with a positive net Bitcoin flow over the past week. — **Over the past few days, some of the largest cryptocurrency exchanges have shared their proof of reserves, revealing how much and what cryptocurrencies they store on behalf of their customers. Binance was among the very first ones to do so, followed by Crypto.com, Bybit, and Huobi.

September 2021 – 🇺🇸 https://www.prnewswire.com/news-releases/cambrian-asset-management-launches-systematic-bitcoin-and-ethereum-trusts-301382419.html name-dropping press release:

September 2021 – https://www.bloombergquint.com/markets/crypto-quant-up-76-breaks-ground-with-vol-cutting-active-trusts they use https://sudrania.com/ as

September 2021 – https://www.tradersmagazine.com/am/large-hedge-funds-to-enter-crypto/ #dream #vision #reverse“We’re eventually going to see a reverse takeover of blockchain and crypto taking over all the other asset classes at institutions and traditional exchanges. We are in the very early days like the cricket like the first innings in baseball.” – David Olsson, global head of institutional distribution at digital financing platform BlockFi.

May 2021 – https://www.icc-usa.com/high-frequency-trading-hft-updates-in-2021/

February 2021 – https://a-teaminsight.com/xilinx-seeks-to-democratise-fpga-in-trading-with-accelerated-algo-framework/

October 2019 – https://globaldigitalfinance.medium.com/crypto-and-the-latency-arms-race-towards-speed-bumps-and-otc-trading-ba8a8d756b0c

April 2019 – https://sites.law.duke.edu/thefinregblog/2019/04/24/high-frequency-trading-comes-to-cryptocurrency/


📍 Recruitment Board 🔑 ** Guest Researcher Area (accessible by invitation only) **🔒

- [LOB Data on S3 🔏](/engineering/integrations/lob-data-on-s3/)
- [Algo-Trading Books 🔒](/research/books/algo-trading-books/)

🔆 External Links

- [**Link-in-Bio for EonLabs**](https://bit.ly/m/eonlabs)
- [**Indeed Company Page**](https://ca.indeed.com/cmp/Eonlabs)
- [**EonLabs.com Official Site**](https://www.eonlabs.com/)

🍁 EonLabs acknowledges that the land on which we live and work is the unceded traditional territories of the xʷməθkʷəy̓əm (Musqueam Indian Band), Sḵwx̱wú7mesh (Squamish Nation), and səlilwətaɬ (Tsleil-Waututh Nation) and we are grateful for the opportunity to do so.