August 28 / Backtesting on Signal Sigma [Measure-Trace Update 2]

Summary of New Features

• Dynamic Mode for Backtests: The screener now supports dynamic mode, automatically including tickers historically available on the simulated date, even if delisted or affected by corporate actions like acquisitions at the present date;

• Filtering: Filters which are applied in the screener now impact backtests, allowing constraints on multiple columns (e.g., volume, industry) in a specific order: starting from the Ticker column and going from left to right through each filter stage;

• AI Anomaly Detection: A new toggle activates AI to detect and smooth irregularities in data, especially for fundamental metrics, ensuring more accurate backtest representation.

Example Instructions for Backtest Setup

1. Set Up Screener:

   • Select dynamic mode to include tickers available on the simulated date;

   • Add filters in order (e.g., ticker column, asset type, then your preferred filters);

2. Define Measurement:

   • Choose a metric to measure (I used the 40-day stochastic oscillator median) and name the backtest;

   • Optionally enable the AI anomaly detection toggle if measuring fundamental data to smooth irregularities;

   • Include secondary studies for signal generation;

3. Run Backtest:

   • Select a start date and launch the backtest.

4. Analyze Results:

   • Review the output;

   • Overlay with relevant assets (e.g., SPY or XLK) to compare with market trends;

   • Use signal analysis to identify patterns;

Upcoming Features

• Signal Trace Type Backtest: Will allow overlaying signals for precise entry/exit points.

• Scheduled Backtests: Option to schedule backtests to run daily for up-to-date results.