SR&ED Claim Information for Ian Moore

Source: Notion | Last edited: 2023-11-23 | ID: a68e0c71-3f7...

---

Name: Ian Moore

Position: Machine Learning Scientist

Duration of Employment for Fiscal Year: January 1, 2023 - April 30, 2023

SR&ED-able Time: 100%

Project Title: Enhancing Trading Algorithms through Dynamic Leveraging, Weighting, and Feature Evaluation

1. Project Description:

• Brief Description: This project focused on developing advanced machine learning techniques to optimize financial trading algorithms, including dynamic leveraging and weighting, and evaluating the effectiveness of time-series features in enhancing performance.

1. Technological Advancements and Challenges:

• Innovative Technologies/Methods: Development of a real-time Kalman filter for local mean estimation and a pre-trained logit curve for optimal leverage determination. Implementation of dynamic weighting of trading intervals using scipy.optimize and evaluation of time-series features for performance enhancement.
• Technological Challenges/Uncertainties: Addressed challenges in reducing maximum drawdown and increasing gains in trading algorithms, and in determining the efficacy of specific time-series features for performance enhancement.

1. Experimental Development and Analysis:

• Experimental Processes/Methods: Utilized advanced statistical methods and machine learning algorithms for dynamic leveraging and weighting. Conducted thorough evaluations to split profit and loss returns for assessing feature efficacy.
• Analysis and Findings: The implementation of these methods resulted in more refined and effective trading strategies, with improved risk management and gain potential.

1. Results and Impact:

• Results Achieved: Developed more sophisticated and efficient trading algorithms with enhanced risk management and profitability.
• Impact on Field/Industry: Contributed significantly to the field of financial machine learning, particularly in the areas of algorithmic trading and predictive analytics.

1. Documentation and Supporting Evidence:

• Types of Documentation: Code repositories, algorithmic models, statistical analysis reports, and data evaluation results.
• Private repositories (contact Eon Labs Ltd. for authorization):
  • https://github.com/Eon-Labs/macro-performance/blob/main/notebooks/tutorials/ensemble/dynamic_leveraging.ipynb
  • https://github.com/Eon-Labs/macro-performance/blob/main/notebooks/tutorials/ensemble/dynamic_weighting.ipynb
  • https://github.com/Eon-Labs/macro-performance/blob/main/notebooks/tutorials/ensemble/eval_feature_split.ipynb

1. Time Allocation:

• 100%

📍 Site Map of SR&ED Claims