Competition Background
BACKGROUND
Introduction
Trading bots are rising in popularity. The purpose of this competition is to create a crypto quantitative trading bot integrated on the most popular crypto exchanges (Binance, Bitstamp, Kraken, etc). The strategy is to recognize price patterns to optimize return and to take advantage of the diversification to reduce the risk associated with a specific crypto currency. The winning strategies can be licensed to become trading bots for cryptos.
About the sponsor
NapBots is a trademark of Napoleon Crypto Group. NapBots is a trading bot platform that designs profitable trading strategies and makes them available in a unique marketplace. It also provides automated execution for users on major crypto platforms through API connection. Alongside NapBots, Napoleon Crypto fully owns and operates, Napoleon AM, one of the first regulated asset managers in France that is dedicated to crypto solutions
OBJECTIVES
Description
You must build a dynamic long only allocation between Bitcoin, Ether, Litecoin, XRP and Tether with the objective to outperform a buy and hold bitcoin strategy.
Your objective is to maximize the Sharpe ratio of your portfolio but also to minimize the drawdown. These 2 metrics will be taken into consideration to select the winner(s) along with the robustness of the code (overfitting measure, clarity of the algo, quality of the comments in the notebook).
Use the tutorial notebook to get started:
Investment Universe
The Investment Universe is composed of the following tickers:
| Ticker | Crypto currency |
| BTCUSD | Bitcoin |
| ETHUSD | Ethereum |
| LTCUSD | Litecoin |
| XRPUSD | XRP |
| USDTUSD | USD Tether |
Data
The only available input data for this competition are:
- BB Live, Open Price, High Price, Low Price, Close Price of the crypto currencies. BB Live refers to the investable index of the respective crypto currencies. Have a look at the investable assets section under this page for more information.
The data can be called using the getTickersCryptoSelector function.
| Ticker | Field | Description |
| BTCUSD | BB Live, OHLC | Bitcoin |
| ETHUSD | BB Live, OHLC | Ethereum |
| LTCUSD | BB Live, OHLC | Litecoin |
| XRPUSD | BB Live, OHLC | XRP |
| USDTUSD | BB Live, OHLC | USD Tether |
In Sample and Out of Sample periods
The in-sample period runs from 1st Jan 2017 to 31st Dec 2020, this is what is given to create your model and given publicly in the Public Leaderboard.
The out-sample period runs at least from 1st Jan 2021 to now but can also have additional data added and created by Alphien. This 'synthetic' data is used to test your model further for its ability to predict future data. This is used in the Private Leaderboard.
The composite period includes all real data available, this is used in the community Leaderboard.
Note: Once your first submission is approved, you can revise it as much as you want.
Benchmark
No benchmark (your implicit benchmark is a cash position in US dollars).
Constraints
| Strategy Constraints | |
| (1) Sum of all weights = 100% |  |
| (2) Max weight per crypto <= 50% | |
| (3) First Signal generated before 2017-04-01 | |
| (4) Sharpe Ratio > 1.5 | |
| (5) No Lookahead bias | |
| (6) Models to be run under 30 seconds | |
| (7) Notebook and payout documentation* | |
*Additional notes:
- All time series data has to be called via the DataFeatures object - no data calls in your payout.
- Payout and its dependencies have to be attached and properly documented.
- Short summary of your model to be written at the start of the notebook -- Summary can include model used, why the model was selected, rationale for any data processing done, parameter selection etc...
- Models will need to be ran under 30 seconds and include a lastN parameter (refer to tutorial notebook)
- Only notebooks that are well documented, code with clear comments will be accepted. The notebook need to explain the rationale of your approach and why you are using the techniques. A lack of clear documentation can induce disqualification even if your version of notebook has been accepted, make your code easily readable and clear to understand, create functions or classes to make your notebook cleaner. Also, put potential references in your notebooks.
Please also have a look at the payout pitfalls page for common issues faced by users.
Strategy transactions
Transaction costs: 10 basis point (bp) of the amount being transacted.
Transaction frequency: The maximum transaction frequency is daily (no intraday trades). All trades are executed at market defined close price set at UTC Universal Time Coordinated.
Transaction signals: Transaction signals must be based on data available at the close-of-business day. Signals are 'executed' the same day following the Lag 0 model convention.
Submissions
You must use Notebook to submit your solution. You can only submit one notebook.
However, there is no limitation in the number of revisions you can make on a submitted notebook, one you have submitted a revision it may take up to 7 days to get it approved or get a feedback on it, if you want to have quick feedback mention as a comment in your submission that you want to use your points for 'Fast Track', check Alphien Points for more details.
If the solution follows a theoretical or academic paper, the participants are encouraged to attach the document with the submission, any relevant document (own paper or other document) to explain the approach used will increase the robustness scoring.
Please visit our Submissions page for all details on how to submit.
SELECTION CRITERIA
The strategies will be ranked based on the weighted arithmetic mean of the following three rankings:
| Performance Metrics | Metric | Weight |
| Normalized Sharpe Ratio | SR | 30% |
| Normalized Maximum Drawdown | DR | 30% |
| Normalized Robustness | R | 40% |
Note:
- The out-of-sample period will not be disclosed.
- If there are multiple versions of the same notebook, only the latest version will be taken into consideration.
Example:
The strategy is ranked according to the different normalized metrics as follow
- Sharpe ratio: 2.0
- Maximum Drawdown: 1.5
- Robustness: 4.45
Its weighted arithmetic mean will be (30% x SR) + (30% x MD) + (40% x R) = (0.3 x 2) + (0.3 x 1.5) + (0.4 x 4.45) = 2.8 to be compared with other strategies.
AWARDS
The total prizes in euros for the winner are:
- Champion: EUR 6,000 or in crypto currencies equivalent at the payment date
- First runner-up: EUR 3,000 or in crypto currencies equivalent at the payment date
- Second runner-up: EUR 1,000 or in crypto currencies equivalent at the payment date
Note: If there is more than one person per team, the payment will be split to each individual team participant equally.
You will also have the opportunity to get your strategy licensed by the Sponsor looking for such crypto strategies.
TIMELINE
Alphathon Starts
Monday 03, May 2021.
Research Ends
Wednesday 17, June 2021 midnight NY Time*.
- New submissions or changes to existing models would not be allowed after the deadline
- Users may still work on the feedback received from the Alphien Team.
- The winners will be announced based on the leaderboard once all the submissions have been processed.
All strategies that have been approved will be open for licensing and paper traded for at least 6 months
*24-Hour extension was granted due to a system issue on the 16th June
PARTICIPANTS
</div>
The competition will be open to any participant (academic or non-academic). You can register as:
- Individual: One person.
- Team: A team is a group of individuals which have registered to participate in the Competition. A Team can be composed of 2 to 3 individuals.
You are not allowed to participate simultaneously as Individual and Team.
ALPHIEN POINTS
- Starting from 12th of May 2021 and every Wednesday until 16th of June 2021, the top 15 users ranked by the selection criteria will earn Alphien Points allowing them to unlock special features for the Crypto Selector Alphathon!
- 1 Fast Track Submission to allow priority review of your submission (once approved an existing submission can take up to 1 week to be reviewed again).
- 1 Extra Submission. Remember that you can only submit one solution by default.
- 1 Special 30 min consultation with a Senior Quant.
- You can request to unlock these features by commenting on the ticket generated after you make a submission or in the comment box of the submission.
- Note that for this competition each user will not receive more than a total of 25 points
- Submissions will have to be approved before midnight eastern time (NY Time - UTC-4hours)
Weekly point rewards
| Ranking | Points |
| 1 | 15 |
| 2 | 11 |
| 3 | 7 |
| 4 | 3 |
| 5 - 15 | 1 |
GUIDELINES
Support tools
- After you have joined the competition, click the "get started" button. You will be redirected to your dashboard. In "Public Notebooks", filter notebooks by competition tag using the "Crypto Selector" tag.
- Read about common payout pitfalls to maximise your chances to win.
- Use the Forum to find help from the community or from Alphien support team.
Tips and advice
- Use AlphienLab to conduct your research.
- Once you have fine-tuned your model, document it in a notebook and submit the notebook for our review.
- We highly recommend you use the Forum as much as you can to get support.
RULES
The Competition will be open to any Participants (academic or non-academic) in the world, but Alphien reserves the right to reject some applicants if terms of the platform are violated. In particular, you must not copy code or intellectual property which is not your own or is not open sourced. You can not participate as a professional if your employer can claim ownership in any part of your work on the Alphien platform; in case of conflict please contact and discuss with the Alphien team. Alphien is open to free scientists who own their intellectual property.
LEGAL
Please refer to Alphien Terms of Use.