Skip to content

Latest commit

 

History

History
14 lines (11 loc) · 449 Bytes

File metadata and controls

14 lines (11 loc) · 449 Bytes

Pipeline:

Model predicts next price for each timestep Generate trading signals from predictions (e.g., "buy if predicted price > current price") Simulate trades over historical data Calculate P&L and risk metrics from the simulated returns Example:

Then compute:

Sharpe Ratio = Mean return / Std dev of returns Sortino Ratio = Mean return / Std dev of downside returns Max Drawdown = Largest peak-to-trough decline VaR = Worst 5th percentile loss