About the Challenge

Welcome to the Joint-University Algo Trading Challenge 2020/21.

Ideate your trading algorithms, test on a robust back-testing platform, and compete to make a name for yourself!

Gain professional industry knowledge and practical skills to prepare yourself as the next-gen algo-trading professional. All undergraduate and postgraduate students from any faculty at participating universities are eligible.

Create teams of 1~4 members with diverse skills (maths, statistics, finance, coding, presentation) and register by 18 Oct 2020.


Eligibility:

All undergraduate and postgraduate students from any faculty at participating universities are eligible.



ideate

1st Round: Ideate

Registered teams of 1-4 students will prepare a trading proposal consisting of:

  • Executive summary
  • Trading idea description
  • Trading logic
  • Implementation detail
  • Team biography

Coding is not required in this round


Submission deadline: 21 Nov 2020

Result announced: 30 Nov 2020


test

2nd Round: Test

Teams will code their trading algorithm and test on ALGOGENE's back-testing engine, then submit the code


Judges will backtest trading algorithms for return, volatility, robustness and practicality to select advancing teams


Submission deadline: 16 Jan 2021

Result announced: 31 Jan 2021


compete

Final Round: Compete

A new seperate set of data will be provided to test the trading algorithm of each team


Teams will present trading plan to judges in a 5-minute pitch session on event day

Judges will select winners in each of these categories:

  • Best Return
  • Best Sharpe
  • Best Strategy Design

Submission deadline: 20 Feb 2021


Judging Criteria

Registration

Form a team of 1~4 members with diverse skills (maths, statistics, finance, coding, presentation) and register by 18 Oct 2020.

Application is closed.


Resources

The dataset provided includes the latest trading records and bid-ask updates by the hour, and covers data from normal trading hours within every trading day from Jan 2010 to Dec 2019. Kindly note that data provided is strictly for trading strategy prototyping purposes only. Final round evaluations will be carried out using a new data set with the same structure but from a different time period. Thus, over-fitting of the data set is discouraged.


Trading Products

  • Foreign Exchange CFDs
  • Commodity CFDs
  • Equity Index CFDs

Datasets

Provide access to required historical data for model development and testing

  • Market data
  • News
  • Economic Statistics

System

Access to ALGOGENE's back-testing engine, cloud simulation environment and relevant technical documents


Parties

Participating Universities:


Event Sponsors:


Media Partners:


Co-organisers:

Results

The final result of the Joint-University Algo Trading Challenge 2020/21 comes out! Congratulation to all winning teams!

Out of sample data: 1/1/2020-31/12/2020




Rank Team Best Return
1 FGF Team (PolyU) +81.99%
2 QuantHeal (CityU) +73.14%
3 AlphaGo (HKUST) +29.85%

Rank Team Best Sharpe
1 MoneyMaker (CityU) 1.7618
2 FGF Team (PolyU) 1.5608
3 QuantHeal (CityU) 1.0849

Rank Team Best Strategy Design
1 AlphaGo (HKUST) Applies NLP algorithms on news to trade NASDAQ Stock Index
2 AlphaBoom (HKUST) Uses LSTM model to trade multiple asset markets on a high-frequency basis
3 CUAmateurs (CUHK) Uses a multi-strategy approach on hourly data to trade FX and Commodities markets



Media Coverage

StartUpBeat

NLP用新聞情緒測美股 科大生奪獎, 信報財經新聞, March 5, 2021

由IBM、瑭明資本等贊助,Algogene、Finbot及Contrendian聯合舉辦的Algo Trading Challenge 2020/21最近公布賽果。香港科技大學一年級生組成的AlphaGo團隊,把人工智能(AI)自然語言處理(NLP)技術,應用於美國新聞情緒分析,作為納斯特指數投資策略的參考數據,奪得「最佳策略設計」首位及「最佳回報」第三名。 ...

StartUpBeat

The finale of Joint-University Algo Trading Challenge 2020/21, InvesTalk, Feb 20, 2021

大學聯校演算交易挑戰盃2020/21 將於2月20日下午2時展開終極決戰!

技術分析, 計量模型, 新聞解讀, 人工智能...

雲端掀戰, 演算爭鋒, 各派學説, 妙着紛呈! 欲知花落誰家, 萬勿錯過!