Timeline for Why does a 95% Confidence Interval (CI) not imply a 95% chance of containing the mean?
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 21, 2023 at 21:12 | comment | added | Ben | @astralwolf: The conditional probability based on observing $\mathbf{x}$ would require a Bayesian posterior calculation, so it is possible to look at things this way in Bayesian statistics (though thte results hinge on prior assumptions). Within classical statistics they want to avoid making prior assumptions about $\theta$ and instead use procedures that tend to do well over all $\theta \in \Theta$. Consequently, within the classical paradigm the assessment of estimators, etc., is based on looking at their behaviour as random variables when the model parameters are fixed. | |
| Mar 21, 2023 at 11:07 | comment | added | astralwolf | I don’t get it. What is the value of saying this next calculation has a 95% chance of containing the pop mean right before calculating it, but then completely rescinding that statement immediately after said calculation takes place. It seems so counterintuitive, what information is being gained here? | |
| Jan 22, 2022 at 8:04 | history | edited | Ben | CC BY-SA 4.0 | added 32 characters in body |
| Jan 22, 2022 at 6:16 | history | answered | Ben | CC BY-SA 4.0 |