{ "fetched_at": "2026-10-19T11:55:12Z", "source_url": "https://www.kaggle.com/competitions/AlgorithmicTradingChallenge/data", "pages": [ { "name": "Description", "content": "

The Algorithmic Trading Challenge is a forecasting competition which aims to encourage the development of new models to predict the stock market's short-term response following large trades. Contestants are asked to derive empirical models to predict the\r\n behaviour of bid and ask prices following such "liquidity shocks".

\r\n

Modelling market resiliency will improve trading strategy evaluation methods by increasing the realism of backtesting simulations, which currently assume zero market resiliency.

", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Background", "content": "

In order driven markets prices are determined by the publication of orders to buy or sell shares. Examples of order driven markets around the world include Nasdaq OMX, London Stock Exchange, Tokyo Stock Exchange and Australian Securities Exchange. In these\r\n markets, participants may submit limit orders or market orders. Limit orders specify the price at which a trader is willing to transact. Orders that do not execute immediately may be stored for later execution in a limit order book. Market orders execute immediately\r\n against orders in the limit order book. An understanding of the dynamic interplay between market and limit orders, and the state of the limit order book is important to traders, exchanges and regulators.

\r\n

The limit order book represents a pool of trading interest over a range of prices. Standing buy (sell) orders with the highest bid (lowest ask) price have the highest probability of execution. The difference between the two best prices is the bid-ask spread.\r\n Competition between market participants ensures that in equilibrium the size of the spread is small.

\r\n

This competition uses trade and quote data (TAQ data). Changes to the state of the order book occur in the form of trades and quotes. A quote event occurs whenever the best bid or the ask price is updated. A trade event takes place when shares are bought\r\n or sold.

\r\n

Liquidity is the ability of market participants to trade large amounts of shares at low cost and quickly. Market resiliency (also known as liquidity replenishment) is the time component of liquidity and refers to how a market recovers after liquidity has\r\n been consumed. Resiliency is very important to market participants, particularly traders wishing to reduce their market impact costs by splitting large orders across time.

\r\n

A liquidity shock is defined as any trade that changes the best bid or ask price. Shocks to liquidity may occur when a large trade (or series of smaller trades) consumes all available volume at the best price. Following a liquidity shock the spread may be\r\n temporarily widened, and/or result in permanent price shifts (see Figure 1). In time, investors replenish the order book with new orders to buy and sell. Market resiliency is that part of mean reversion where the spread and depth partially or completely revert\r\n to former levels following a liquidity shock.

\r\n

Figure 1     Liquidity Replenishment after a Shock

\r\n

The figure depicts a limit order book in event time. The event types are either ‘quote’, whereby the best bid or the ask price is updated, or ‘trade’, whereby shares are transacted.\r\n The red (green) coloured section represents standing orders to sell (buy). A liquidity shock is defined as a trade that changes the best bid or ask. Liquidity replenishment is the stock price mean reversion following a shock.

\r\n

\"Liquidity

\r\n

The aim of this competition is to determine the relationship between recent past order book events and future stock price mean reversion following shocks to liquidity. The extant literature suggests a number of factors to consider in the development of a\r\n liquidity replenishment model, including:

\r\n\r\n

The data includes variables relating to the above factors.

", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "

Performance evaluation will be conducted using root mean square error. For each prediction, RMSE will be separately calculated for the bid and ask at each time step following a liquidity shock. The winning model will be the one with the lowest cumulative\r\n RMSE across the entire prediction set.

", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Rules", "content": "\r\n\t\r\n

One account per participant

\r\n

You cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.

\r\n

No private sharing outside teams

\r\n

\r\n Privately sharing code or data outside of teams is not permitted.\r\n It's okay to share code if made available to all participants on the forums.\r\n

\r\n

Team Mergers

\r\n

\r\n Team mergers are allowed and can be performed by the team leader. In order to merge, the combined team must have a total submission count less than or equal to the maximum allowed as of the merge date. The maximum allowed is the number of submissions per day multiplied by the number of days the competition has been running.
\r\n

\r\n\r\n

Team Limits

\r\n

There is no maximum team size.

\r\n

Submission Limits

\r\n

You may submit a maximum of 2 entries per day.

\r\n

You may select up to 5 final submissions for judging.

\r\n

Competition Timeline

\r\n

Start Date: 11/11/2011 6:05 PM UTC

\r\n

Merger Deadline: None

\r\n

Entry Deadline: None

\r\n

End Date: 1/8/2012 11:59 PM UTC

\r\n\t\r\n
    \r\n
  1. Winning participants are required to provide code, as well as a description of their algorithm in document form.\r\n
  2. All participants are encouraged to provide code and documentation of their methods.\r\n
  3. Capital Markets CRC Limited retains the right to not award a prize if it is determined that a player has violated any of the competition rules.\r\n
  4. The challenge organisers should be notified if any article is written using results or data from the Algorithmic Trading Challenge.\r\n
  5. An article will be written summarising the results from the Algorithmic Trading Challenge. By submitting an entry into the challenge you agree that a discussion of the method may appear in this article.\r\n
  6. Participants will not use the data for commercial purposes.
  7. Participants will not use the data for purposes unrelated to the competition.\r\n
  8. Participants will not use data other than that provided to estimate their model.\r\n
  9. If necessary, the competition host may modify the training and/or testing dataset to address any problems/issues that come up during the competition.\r\n
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Prizes", "content": "

The competition winner will receive:

\r\n\r\n

There are also two milestone prizes of $1,000 each, which will be awarded to the competition leader on November 30 and December 22, 2011, respectively.

", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Help", "content": "

Please write any questions or comments on the user forum.

", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "data-description", "content": "

NOTE: The data files were all updated on November 11, 2011. Please download them again if you downloaded them before that date.

\r\n

Recent trade and quote data from the London Stock Exchange (LSE) is provided. The preprocessed dataset comprises observations of the limit order book before and after a liquidity shock (a trade that results in widening of the bid-ask spread).

\r\n

Table 1 shows the data schema provided for each liquidity shock.

\r\n

\"Data

\r\n

Table 2 shows information about each data field.

\r\n

 \"Data

\r\n

Contestants are asked to submit predictions based on a test dataset containing 50,000 rows. A larger dataset is provided for training purposes.

", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "SubmissionInstructions", "content": "

For each observation a participant should provide 100 numbers describing the next 50 bid and ask prices following a liquidity shock. There should be 50,001 rows in total (50,000 entries plus a header row) and 101 columns for each row.

", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Custom", "content": "

The competition is organised by the Capital Markets Cooperative Research Centre (CMCRC - www.cmcrc.com), a group of researchers, universities and industry partners with a long history of applying academic research techniques to solving commercial problems.\r\n The CMCRC attracts high calibre Australian and international researchers to work together with industry partners to develop new and innovative technologies for the capital markets domain.

", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Custom", "content": "

This competition uses recent trade and quote data from the London Stock Exchange (LSE).

\r\n

The LSE is one of the world’s largest and most international organised markets with more than 3,000 companies listed from over 60 countries. The market is structured essentially as an electronic communication network, with multiple dealers and an electronic\r\n order display book. In 2009, on the LSE’s limit order book alone, over 600,000 trades worth more than £4 billion (US$ 6.5 billion) were traded each day.

\r\n

The stocks trade on weekdays between the hours of 08:00 and 16:30. The LSE trading engine matches buy and sell orders on a price/time priority. The market is an electronic trading environment with both market makers and anonymous orders. The LSE has registered\r\n market makers, whose purpose is to enhance liquidity and facilitate orderly trading. Although all stocks are eligible to participate, the focus is on medium and low liquidity stocks. In these agreements, a member firm agrees to maintain two-sided quotes on\r\n the market’s central order book of at least a minimum size throughout the trading day. As a hybrid market, the limit order book may contain both public orders and those of registered market makers.

", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false } ] }