Quote of the day

“A statistical model is usually taken to be summarized by a likelihood, or a likelihood and a prior distribution, but we go an extra step by noting that the parameters of a model are typically batched, and we take this batching as an essential part of the model.”

3 thoughts on “Quote of the day

  1. Batched is perhaps a botched term for this – discerning which parameters are common versus arbitrary by observations/groups in addition to which parameters are specified as being random draws from a common distribution (ala exchangeability formalisms) would arguably also be worthwhile.

    For instance, for a multilevel model for repeated studies it would be a common start to have treatment group variances common within the studies, the variances arbitrarly different between studies and with random treatment effect parameters _drawn_ independently by study from a common distribution.

    But then when I used that vocabulary in my thesis the examiners felt it worth while to point out that such nonstandard vocabulary would likely distract the interest of potential readers.

    K?

Comments are closed.