volume 13, issue S337, P344-345 2017
DOI: 10.1017/s174392131700850x
View full text
|
|
Share
Andrei P. Igoshev, Frank Verbunt, Eric Cator

Abstract: AbstractIf the observed parallax ϖ′ has a gaussian measurement error σ, there is a 68% probability that the actual parallax ϖ is in the range ϖ′ − σ < ϖ < ϖ′ + σ (the frequentist approach). The probability distribution within this range is not known from ϖ′ and σ alone, and in particular, we cannot state that the most probable distance D is given by D = 1/ϖ′. To obtain a probability distribution, we need to know or assume a distribution of pulsar distances. Similar assumptions are also required to estima…

expand abstract