LBVs and Statistical Inference

Abstract

Smith and Tombleson (2015) asserted that statistical tests disprove the standard view of LBVs, and proposed a far more complex scenario to replace it. But Humphreys et al. (2016) showed that Smith and Tombleson's Magellanic "LBV" sample was a mixture of physically different classes of stars, and genuine LBVs are in fact statistically consistent with the standard view. Smith (2016) recently objected at great length to this result. Here we note that he misrepresented some of the arguments, altered the test criteria, ignored some long-recognized observational facts, and employed inadequate statistical procedures. This case illustrates the dangers of uncareful statistical sampling, as well as the need to be wary of unstated assumptions.

References in corpus (1)

• Publication bias and the canonization of false facts

Cited by in corpus (5)

• Massive runaways and walkaway stars
• The possible detection of a binary companion to a Type Ibn supernova progenitor
• Gaia TGAS search for Large Magellanic Cloud runaway supergiant stars:Candidate hypervelocity star discovery, and the nature of R71
• How long can LBVs sleep? A long-term photometric vaiability and spectral study of the Galactic candidate luminous blue variable MN112
• Radiation-Driven Stellar Eruptions