DST Market Mind - "High Frequency" style trading algo based on the Dempster-Shafer fusion theory in C# using the Interactive Brokers API.

Dempster-Shafer Theory (DST): data fusion method based on mass functions (Code Sample)

• Event based automated trading strategy

• Use's Dempster’s rule of combination to fuse information from multiple independent sources

• User/Subject Matter Experts select parameters used to generate mass functions.

• Mass function 1: market microstructure(spread, order book depth, queue, short interest, etc.)

Elements of Market Depth evaluated: Bid Ask Depth and Thickness Spread > Average Spread Bid Price Improvement Degree in which the Bid price improves Ask Price Improvement Degree in which the Ask price improves Ask Size Improvement vs. Average Ask Size Improvement Bid Size Improvement  vs. Bid Ask Size Improvement

• Mass function 2: Event based Moving Average (EMA) shrinkage and crossovers

• Assigns weights solely on what information is available. Not hindered by missing information.

• Utilize's Interactive Brokers TWS C# API

• Fully functional GUI

• Estimated refresh rate: 5-15 seconds

Developed by

Randall Campbell and Francisco Hernandez @ The Illinois Institute of Technology - Stuart School of Business