Timeline for What is a generalized confidence interval?
Current License: CC BY-SA 3.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 26, 2017 at 14:59 | comment | added | whuber♦ | Thank you for the edit. I still do not see anything in it that contradicts my definition (2). Indeed, the theorem you quote is a strong law of large numbers following from (2). The only difference I detect between the two settings is that I see no reason to require that the sequence $s$ consist of distinct statistical models; it's only necessary that the samples from each one be independent. | |
| May 26, 2017 at 14:44 | comment | added | Stéphane Laurent | Dear whuber, I've edited my post to include a theorem which makes the definition clearer. Though I personally still need some time to digest it. | |
| May 26, 2017 at 13:40 | comment | added | Stéphane Laurent | This is the only definition I have found. The other "definition" I found is the definition of the construction of a GCI. But maybe I missed something. I'll check again when I'll have the opportunity. | |
| May 26, 2017 at 13:25 | comment | added | whuber♦ | Could you quote a definition from the textbook? | |
| May 26, 2017 at 13:02 | comment | added | Stéphane Laurent | Sure. But if I had a clear definition I would not have opened this question ^^ | |
| May 26, 2017 at 12:23 | comment | added | whuber♦ | Although a construction could help illustrate the GCI, what we need is a sufficiently clear definition. | |
| May 26, 2017 at 10:56 | comment | added | Stéphane Laurent | Hi whuber. I can provide the definition of the construction of a GCI, but I'm not sure this will help. I'll do it later. | |
| May 14, 2017 at 22:05 | comment | added | whuber♦ | Thank you. In that case, your quotation does not provide a sufficiently clear definition. Could you provide any information to make it less ambiguous? | |
| May 14, 2017 at 18:17 | comment | added | Stéphane Laurent | Hello whuber. It is not true that a generalized confidence interval is a confidence interval. Hence Property 2 should not imply Property 1. I should have mentioned this point in my post. | |
| May 14, 2017 at 18:01 | history | answered | whuber♦ | CC BY-SA 3.0 |