Contrarian Trading Against an AI News Sentiment Signal
Summary
The document presents a contrarian idea: take the opposite side of a machine learning system’s probability forecast derived from financial news. If its signal indicates a likely rise, the strategy sells; if it indicates a likely decline, it buys. The example retrieves a probability series for Bitcoin and uses recent highs and lows plus 0.6 and 0.4 thresholds to trigger short and long entries.
It describes the method as a demonstration of calling an external data source and evaluating it in a backtest. The published settings specify hourly Bitcoin trading on Bitfinex over a historical interval, but the document provides no readable performance results or evidence that the contrarian premise is profitable. It also offers no explanation of the model, forecast calibration, costs, or risk controls. The source explicitly cautions against using the example for live trading, so its claims about the signal and reversal should be treated as unverified.
Key ideas
- The strategy reverses the direction of a machine learning forecast based on financial news.
- A probability above 0.6 and a recent high trigger a short entry in the example.
- A probability below 0.4 and a recent low trigger a long entry.
- The example demonstrates external data retrieval and historical testing, not verified profitability.
- The document warns against live deployment.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.