We are excited to announce the launch of our new forums! You can access it forums.aavso.org. For questions, please see our blog post. The forums at aavso.org/forum have become read-only.
In his light curve analysis book, Foster says this about autocorrelation:
It can happen that different noise values are in fact correlated. It often happens, for instance, that the correlation between two different noise values depends on the time difference be- tween them. This has a profound effect on the statistical be- havior of analysis of the data. In such cases we say that the noise, when treated as a time series, exhibits autocorrelation.
As has been emphasized, the noise in light curve data are generally assumed not to show autocorrelation. Data from other fields, e.g. geophysical data, exhibit autocorrelation so often that understanding its statistical impact is essential for reliable analysis.
No but I've captured it here: https://github.com/AAVSO/VStar/issues/349
As you can see from the links there, this has been on the TODO list for quite awhile.
The original SourceForge issue linked from there has some detail (e.g. see https://www.itl.nist.gov/div898/handbook/eda/section3/eda35c.htm), but if anyone has suggestions re: algorithm specifics, feel free to add here or to the GitHub issue.
In his light curve analysis book, Foster says this about autocorrelation:
David
Well, it's not a magic wand, but useful tool for quasi-periodicity...
https://www.aavso.org/vsots_rvtau
Del