Skip to main content
11 events
when toggle format what by license comment
Oct 1, 2015 at 5:36 history edited user36160 CC BY-SA 3.0
added 4 characters in body
Sep 30, 2015 at 19:43 comment added user36160 @jlimahaverford - It reads now "This gives a rule such that, if applied repeatedly to similar problems, the 95% confidence interval will contain the true parameter value in 95% of the cases."
Sep 30, 2015 at 19:39 comment added jlimahaverford hmm, I am not seeing it corrected.
Sep 30, 2015 at 19:37 comment added user36160 @jlimahaverford - aggree and corrected - Thanks.
Sep 30, 2015 at 19:35 history edited user36160 CC BY-SA 3.0
added 2 characters in body
Sep 30, 2015 at 19:33 comment added jlimahaverford I was not talking about your description of credible intervals I was talking about confidence intervals. I'm now noticing that in the middle of your paragraph on confidence intervals you start talking about credible again, and I think this is a mistake. The important idea is this "If this were the true value of the parameter, what is the probability that I would draw a sample this extreme or more. If the answer is greater than 5% it's in the confidence interval."
Sep 30, 2015 at 19:22 comment added user36160 @jlimahaverford - The description is correct as is yours. To make the link to what you describe, I added "This gives a rule such that, if applied repeatedly to similar problems, the 95% credible interval will contain the true parameter value in 95% of the cases."
Sep 30, 2015 at 19:19 history edited user36160 CC BY-SA 3.0
adapted to cover comment
Sep 29, 2015 at 20:58 comment added jlimahaverford The description of the confidence intervals is not correct. The "95%" comes from the probability that a sample from the population will product an interval that contains the true value of the parameter.
Sep 29, 2015 at 19:17 review Late answers
Sep 29, 2015 at 20:58
Sep 3, 2015 at 16:20 history answered user36160 CC BY-SA 3.0