Search papers, labs, and topics across Lattice.
This paper identifies look-ahead-freedom as a formal property related to temporal non-interference, addressing the pervasive issue of look-ahead bias in backtesting and agentic trading pipelines. By developing a pipeline calculus that separates data availability from its reference time, the authors provide a sound and decidable type-and-effect system for a value-independent fragment of the problem. The proposed method not only scales linearly but also outperforms existing empirical detectors by reliably identifying leaks that they miss.
Look-ahead-freedom reveals a fundamental flaw in backtesting that could undermine the reliability of machine-learning evaluations in trading systems.
Look-ahead bias (using information from after a decision epoch to make the decision at that epoch) is the dominant way a backtest or a machine-learning evaluation flatters a system that will disappoint in deployment. The field manages it with construct-specific recipes and empirical detectors, which are sound only channel by channel and certify nothing by their silence. We show that look-ahead-freedom is a formal property in disguise: fixing an epoch, the demand that the future not influence the present is temporal non-interference over a time-indexed information lattice. From this identification we develop a pipeline calculus separating a datum's availability from its reference time, and settle the problem's boundary. Where availability may depend on data values, look-ahead-freedom is undecidable (indeed Pi-0-1-hard): leakage is recursively enumerable but freedom is not. On the value-independent fragment (covering windowing, resampling, joins, point-in-time and vintage reads, and agentic retrieval) we give a type-and-effect system that is sound and decidable in linear time. An artifact confirms the theory: the check scales linearly, an independent oracle witnesses no leak in any accepted pipeline, and the checker catches every planted leak that differential and tiling detectors miss.