New method accurately predicts chaos in complex systems from time series.
Lyapunov exponents are numbers that help us understand how stable or chaotic a system is. Scientists have found a way to estimate these numbers from data collected over time. By looking at how nearby points in the data spread apart, they can figure out the largest Lyapunov exponent. This method works well for simple chaotic systems, like the ones they tested. But for more complex systems, they need longer data and have to be careful about which points to compare.