Page tree
Skip to end of metadata
Go to start of metadata

The Order Book AI tool uses a Deep Learning based heuristic matching algorithm to search through vast amounts of market data books and find those that match the input pattern. The user can leverage the search to find when a pattern occurred in the past and what the behavior of the market was after the pattern occurred to develop market intelligence.

Order Book Search

The front end of the order book search requires the following inputs:

  1. the contract
  2. the start time of the 20-second time window of market events that you want to use to find similar past market patterns.



1.  Contract Product: Select the CME Group futures product from the drop down list.

2. Month, Year: Specify the contract month and year you want to search.

3. Duration: Select the starting time in terms of hours, minutes and seconds. Please note, the model has been trained to match 20 second windows of data.  Therefore, the end time is automatically chosen to be 20 secs after the start time. 

4. Click Search to start the search on historical data back to January 4, 2016.


Search Results:
The search results are shown as an input graphic followed by 3 matched outputs (historical book trends) which are graphical and indicate the matched contract and time of the match as shown below.

Input Graph:

Output Graphs: There are three output graphs.  You can iterate through the Output graphs using the and buttons.


Graph Representation: