/ASA

Algo Soc Agent

Primary LanguagePython

📈 AlgoSoc-Sessions

Build Status Java Version Github Repo Size

Pedlar 2.0 Demo's

Pedlar 2.0 Repo

Content

  • data:
    • Contains all the csvs used for backtesting.
  • intro_agents:
    • Contains 3 simple intro agents demonstrating key pedlar functionality.
  • agents:
    • Contains the agents produced during the sessions. Frequently redesigned with improvements and optimisations.

Next Steps:

For agents:

  • Decision tree based agent
  • Linear model based agent
  • NN based agent
  • RNN based agent
  • RL based agent

For finding parameters:

  • Explicit search
  • Random search
  • Grid search
  • Gradient-based search
  • Evolution-based search
  • Sharpe Optimisation

Combining multiple signals & agents:

  • Ensemble of agents
  • K-Armed bandits

Risk Control:

  • Time dependent behaviour (e.g. varying behaviour around market open/close)
  • Real time statistics
  • Add post-analysis of trades made.

icats_logo

About us:

Algo soc website

Algo soc slack

Our trading platform Pedlar:

Imperial pedlar server → (requires access to Imperial Wifi Network)

Pedlar

You can contact us at: algo.trade@imperial.ac.uk

icats_logo