Original title: A Missing Tool for Calculating Auto/Cross-correlation Function under Nonuniform Sampling Observations
Chen-Ran Hu, Yong-Feng Huang, Jin-Jun Geng, Orkash Amat +9 · September 2026
Astronomers study how objects in space change over time, but their telescopes often take data at irregular intervals. This uneven schedule breaks standard formulas used to track repeating signals or time delays between signals, forcing scientists to use workarounds like guessing missing points or grouping data. These workarounds create unknown biases and lack reliable error estimates. To fix this, researchers created a new method for uneven data. Instead of guessing missing points, the technique scores data pairs by using time-interval weights and penalties for how poorly measurements align in time. Computer simulations then determine confidence bands and supply a full error budget covering flux uncertainties and irregular sampling. Tests showed the tool beats older approaches across both strict cycles and complex repeating patterns. Real data tests picked out repeating stellar changes, measured delays in an active galaxy, and validated delays in a lensed quasar.
Imagine trying to find a friend's walking rhythm from a video where the camera only records a few random seconds every minute. You cannot just fill in the missing seconds with guesses because you would invent steps that never happened. Instead, you look at the actual recorded pieces and measure how the time gaps between them relate to the positions you see. In this paper, the irregularly sampled time series are like those random video clips, and the method uses the real gaps and misalignments to compute correlations without inventing data.
Most astronomical data arrives at irregular, unpredictable times—clouds, telescope schedules, and orbital mechanics all break the neat intervals we expect. Until now, researchers had to fill those gaps with guesswork, introducing hidden biases and no way to know how wrong they might be. This method gives them a rigorous way to detect patterns and measure time delays directly, which matters for everything from understanding black hole feeding to decoding fast radio bursts.
The abstract does not specify computational cost or scalability to extremely large datasets. It also does not compare against all possible ad-hoc methods or discuss performance when sampling is extremely sparse (e.g., fewer than a handful of points). The method's reliance on Monte Carlo simulations may be computationally intensive, and the abstract does not quantify the improvement over interpolation in terms of bias reduction or error accuracy.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
A Missing Tool for Calculating Auto/Cross-correlation Function under Nonuniform Sampling Observations — Chen-Ran Hu, Yong-Feng Huang, Jin-Jun Geng et al.
Abstract licence: CC0
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.