A new macro sentiment driven approach to forecasting US equity index futures
-Highlighting the growing role of real time news analytics in systematic investing.
Alexandria Technology leverages global macroeconomic sentiment data produced derived directly from Dow Jones news articles as they are published. Dow Jones generates more than twenty five thousand US economy related articles each month, creating a deep and continuously updated corpus of information for signal extraction across both short term and longer horizon trading strategies.
Each article is machine classified by entities and economic topics, then assigned a probabilistic sentiment score that reflects the tone of the coverage. The system further tags articles by macroeconomic event type, including trade balance, inflation, interest rate policy, employment, and other core drivers of market expectations. This structure allows sentiment to be aggregated into daily topic level signals for the United States.
The model is trained using a rolling fifteen year in sample window, with parameters re estimated annually to reflect evolving market dynamics. Performance is then evaluated strictly out of sample. According to the results, the strategy achieved an out of sample Sharpe ratio of one point six, a level that places it among the stronger systematic macro signals tested on US equity index futures over the past decade.
The findings reinforce the idea that macroeconomic narrative, when measured quantitatively and at scale, can offer persistent information about future market returns. The firm notes that the framework is flexible enough to support strategies ranging from intraday positioning to longer-term tactical allocation.
The research also underscores the commercial value of structured news data as an alternative signal source, particularly as traditional macro indicators are released with delays and revisions.
More details on the methodology and results are available upon request through Alexandria Technology at requests@alexandriatechnology.com



