Give it a thumbs up if it helps you!
Powered By Longer - Standing on the shoulders of giants
Reprint please indicate the source: Longer2048's CSDN blog
illustrate:
Recently, I took the time to summarize, simplify and organize the information on the Internet, and sorted them in chronological order. Welcome to leave a message to add relevant information, and I will update it from time to time.
Remember to like and bookmark.
------------------------------------------ Kaggle Competition ------------------------------------------
in progress:
2022.10.08 (Ongoing): JPX Tokyo Stock Exchange Prediction ($63,000/1372 teams)
简述：Explore the Tokyo market with your data science skills.
Solution: None;
finished:
2022/07/20：Ubiquant Market Prediction($100,000/2893 teams)
Brief description: Make predictions against future market data. Jiukun Investment. Use machine learning algorithms to predict the future rate of return of stocks;
Data: Contains features of real anonymized historical data from thousands of investments. (Rumored data leak: additional data is allowed. The contestants used publicly available Ashare historical data. The feature latitude is 300 dimensions, including training data within 5 years. The id is given as a feature. Many players can match the historical trend. get a high score)
Evaluation: Pearson correlation coefficient;
Scheme: [ 17th ];
2022/03/04：G-Research Crypto Forecasting($125,000/1946 teams)
Brief description: Use your ML expertise to predict real crypto market data. Virtual currency competition, use your machine learning expertise to predict the short-term returns of 14 popular cryptocurrencies.
Data: Contains historical transaction information of some encrypted assets, such as Bitcoin and Ethereum.
Evaluation: Pearson correlation coefficient;
Scheme: [ 18th ];
2022/06/11：Optiver Realized Volatility Prediction($100,000/3952 teams)
Brief description: Apply your data science skills to make financial markets better. Build models to predict short-term volatility of hundreds of stocks in different industries.
Data: Stock market data related to the actual execution of financial market transactions, including order snapshots and executed trades. (Rumored data leak: Some players discovered and used the leaked information in timing, which made the game meaningless)
Assessment: RMSPE;
Model: LightGBM, deep learning model;
Scheme: [ 1st ] - [ 2nd ] - [ 3rd ] - [ 12th ] - [ 20th ] - [ 23rd ] - [ 32nd ] - [ 37th ];
2021/08/24：Jane Street Market Prediction($100,000/4245 teams)
Brief description: Test your model against future real market data. Use historical data to classify potential trading opportunities.
Data: 129-dimensional anonymous data;
Evaluation: two classifications* transaction income;
Model: Deep Learning Autoencoder;
Scheme: [ 1st ] - [ 3rd ] - [ 23rd ] - [ 39th ] - [ 44th ];
2019/08/06：Two Sigma: Using News to Predict Stock Movements($100,000/2927 teams)
简述：Use news analytics to predict stock price performance.
Scheme: [ 7th ];
2017/03/02：Two Sigma Financial Modeling Challenge($100,000/2063 teams)
Brief description: Can you uncover predictive value in an uncertain world? Provide exceptionally accurate predictions in an uncertain world.
Data: anonymous features related to time-varying values of financial instruments;
Evaluation: R-squared regression;
model: linear model;
Scheme: [ 7th ] - [ 10th ] - [ 12th ];
2016/06/27：The Winton Stock Market Challenge($50,000/829 teams)
Brief description: Join a multi-disciplinary team of research scientists. Predict intra-day and end-of-day returns given historical stock performance and many masked characteristics.
Data: characteristics of stocks at the minute level;
Evaluation: weighted mean absolute error;
model: linear regression;
Scheme: [ Solution Sharing ];
2013/10/02：The Big Data Combine Engineered by BattleFin($18,500/424 teams)
简述：Predict short term movements in stock prices using news and sentiment data provided by RavenPack.
Solution: None;
2012/01/09：Algorithmic Trading Challenge($10,000/111 teams)
Brief description: Develop new models to accurately predict the market response to large trades. Predict the market's response to large transactions.
Data: Using Trade and Quote Data (TAQ Data), recent trade and quote data from the London Stock Exchange (LSE).
Evaluation: RMSE regression;
Model: linear regression, KNN and MLP fusion;
Scenario: [ discussion ];
material:
Jane Street Market Prediction and Optiver Realized Volatility Prediction finally ranked the first solution [ zhihu ]
Jane Street Market Prediction Champion Project Experience Sharing[ link ]
Give it a thumbs up if it helps you!
Powered By Longer - Standing on the shoulders of giants
Reprint please indicate the source: Longer2048's CSDN blog