Project Overview
The project is aimed at investigating how credit derivatives are leveraged by mutual funds by studying their routine filings to the US Securities and Exchange Commission
Methodologies
-
Using Python and Perl to extract the reports from the SEC archives and then using NLP to analyze the data
-
With the help of NLP, we will try to understand the complex sentences in various reports and convert it into structured data
-
Conduct time series analysis using auto-encoders or LSTM-RNN to derive key insights from the extracted data
Timeline
Research & Project Plan
- Research on Credit Default Swap
- Literature Review
- Familiarization with Perl and RegEx
September'18
Testing
- Crawling data from the SEC EDGAR
- Extracting the data from the reports crawled
- Testing Natural Language Processing Methods
October - December'18
Presentation and Deliverables of Phase 2
- Preliminary Implementation
- Detailed Interim Report
- Initial Presentation Deck
January - February'19
Deliverables of Phase 3 and Final Presentation
- Finalised Tested Implementation
- Final Report
- Final Presentation
- Project Exhibhition
March - April'19
Team
-
Dr. Luo Ruibang
Assistant Professor (HKU)
Supervisor
-
Tiffany JunYu Wang
PhD Candidate (HKUST)
Co-Supervisor
-
Saripalli Varun Vamsi
Year 4, Computer Science
-
Tarun Sudhams
Year 4, Computer Science
Findings
-
Dashboard
-
CDS Database
-
Credit Default Swap Search
Documentation
-
Project Plan
-
Interim Report
-
Data Preporcessing Report
-
Natural Language Processing Report