8

Statistical significance is denoted to occur when:

In statistical hypothesis testing, statistical significance (or a statistically significant result) is attained when a p-value is less than the significance level (denoted α, alpha)

(According to Wikipedia). While I understand that there is such a thing as Bayes factors (equivalent to alpha values in Frequentist statistics). It seems that the definition of statistical significance is embedded in Frequentist statistics exclusively. Can anyone confirm this?

Nuclear241
  • 810
  • 2
    I think this question already has an answer http://stats.stackexchange.com/questions/197555/does-bayesian-statistics-have-no-concept-of-statistical-hypothesis-testing#comment374529_197555 . Otherwise my advise would be to spend some time learning about what the excerpt you have means (and what a p-value is) and how Bayesian and frequentest models are estimated (you should learn the math). Once you know these things, the answer becomes glaringly obvious. Until then you have to rely on the rants of random people on the internet, which is not conducive of anything really. – Zachary Blumenfeld Feb 28 '16 at 01:32
  • This is Wikipedia and various other sources. This seems to be an issue with terminology. Sorry - as you could tell, I'm new to Bayesian stats and stats in general. I learn a lot when I get great direct responses here. –  Feb 28 '16 at 02:30
  • 1
    Bayes factor equivalent to an alpha value? Definitely not. – conjugateprior Feb 28 '16 at 03:04
  • 2
    You don't have to apologize. It's a meritable question and the answer is yes (for the most part, there are some exceptions). The issue is that your question reveals confusion about the meaning of statistical significance and the interpretation of Bayes factors/posterior probabilities. So to explain my answer I would have to go over the definitions and formulations of p-values, posterior distributions, etc... which would be a very long answer. – Zachary Blumenfeld Feb 28 '16 at 03:05
  • 1
    Right now the better questions for you to ask are "what is a p-value/statistical significance really?", "what do Bayes Factors tell us?", "what are credible intervals and how do we interpret them?"...etc. Once you understand these concepts you will easily be able to answer this question yourself. – Zachary Blumenfeld Feb 28 '16 at 03:06

0 Answers0