[Q & A]Bayesian interval significance?

1449

收藏 2014-06-30

As far as I know, if a confidence interval of a parameter covers 0, then this parameter is said to be statistically insignificant at this level. I am wondering whether this is the case for a Bayesian interval. In other words, if a Bayesian interval of a parameter covers 0, then can this parameter said to be statistically insignificant?

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

全部回复

Lisrelchen

2014-6-30 22:53:50

In a Bayesian framework, you talk about posterior probabilities rather than significance. So if an equal-tailed interval for a parameter is centred around positive values but still covers 0, then you can say there's over a 2.5% probability that the parameter isn't actually positive.

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

扫码加我 拉你入群

扫码加我 拉你入群

分享

扫码加好友，拉您进群

扫码加我拉你入群

扫码加我拉你入群