https://github.com/sjtechdev/hft_48_hour
ML-based HFT algorithm built and backtested in 48 hours. See Final Report for details (the report was also written within that time).
https://github.com/sjtechdev/hft_48_hour
Last synced: 9 months ago
JSON representation
ML-based HFT algorithm built and backtested in 48 hours. See Final Report for details (the report was also written within that time).
- Host: GitHub
- URL: https://github.com/sjtechdev/hft_48_hour
- Owner: sjtechdev
- Created: 2016-06-19T21:50:09.000Z (over 9 years ago)
- Default Branch: master
- Last Pushed: 2016-06-19T21:54:18.000Z (over 9 years ago)
- Last Synced: 2025-04-03T17:01:30.355Z (10 months ago)
- Language: R
- Homepage:
- Size: 661 KB
- Stars: 5
- Watchers: 2
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.txt
Awesome Lists containing this project
README
Hi! Here's the readme file for the project.
HFT_FINAL.pdf - Final Report (Detailed README)
apply_fs.R - removes elminated feature from all the data
classifier.R & classifier_testing.R - contains classifier training & evaluation
extract_preds.R - make predictions for execution algos
fe.R - feature engineering file. Note that I played around with data used in hw3 in this file originally, so its creation time was before the start of my exam, but I did not start coding for the exam until the start.
fs.R - feature selection file
trading_engine_*.R - contains trading simulation
Best,
Shreyas