Abstract
Abstract
Recent studies have found evidence for parity violation in the BOSS spectroscopic galaxy survey, with statistical significance as high as 7σ.
These analyses assess the significance of the parity-odd four-point correlation function (4PCF) with a statistic called X
2.
This statistic is biased if the parity-even eight-point correlation function (8PCF) of the data differs from the mock catalogs.
We construct new statistics X
2
×, X
2
null that separate the parity violation signal from the 8PCF bias term, allowing them to be jointly constrained.
Applying these statistics to BOSS, we find that the parity violation signal ranges from 0 to 2.5σ depending on analysis choices, whereas the 8PCF bias term is ~ 6σ.
We conclude that there is no compelling evidence for parity violation in BOSS.
Our new statistics can be used to search for parity violation in future surveys, such as DESI, without 8PCF biases.