backtest-diagnose

Diagnose zero-trade backtests by analyzing signal engine logic and data filtering to identify entry condition failures. The skill classifies issues into runtime errors, logic bugs, or data-source problems and provides targeted code edits to repair signal generation.

Algorithmic trading backtest failures with NaN propagation stem from data-source problems or logic bugs. The skill diagnoses these runtime errors by analyzing artifacts and source code, then applies precision repairs to signal engines to prevent NaN propagation.

Fixing late entry signals requires diagnosing logic bugs within the signal engine and data filtering rules. The skill repairs failed algorithmic trading backtests by providing targeted code edits and verifying results through automated backtest reruns.

Automated diagnosis fixes runtime errors in algorithmic trading backtests by classifying issues using a structured taxonomy. The skill analyzes backtest artifacts and source code to provide precision repairs and ensures fixes pass AST syntax checks and backtest reruns.

Validate backtest fixes by running automated backtest reruns and AST syntax checks on the repaired signal engine code. The skill ensures backtest integrity and performance by iteratively validating that the targeted code edits resolve the diagnosed issues.