Timeline for Do uncalibrated "probability" predictions satisfy Kolmogorov's axioms?
Current License: CC BY-SA 4.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 3 at 19:41 | comment | added | kjetil b halvorsen♦ | The binary assumption plus independence is enough to determine the probabilities completely. Without independence, see (especially answer by whuber) stats.stackexchange.com/questions/645951/… | |
| Jun 3 at 17:18 | comment | added | Dave | Why is the independence assumption important? | |
| May 20 at 17:49 | vote | accept | Dave | ||
| Sep 20, 2023 at 16:38 | comment | added | kjetil b halvorsen♦ | @Dave: As long as independence is assumed, there is no difference | |
| Sep 13, 2023 at 15:28 | comment | added | Dave | I agree that it is done if we only care about each individual Bernoulli distribution. However, why don’t we care about the entire collection of predictions? | |
| Sep 13, 2023 at 15:19 | comment | added | kjetil b halvorsen♦ | The point is maybe that no checking needs to be done, as it is already done! | |
| Sep 13, 2023 at 15:04 | comment | added | Dave | That makes sense. However, something is still troubling about this. Is the point of checking the Kolmogorov axioms that we check them for each Bernoulli distribution with fitted/predicted probability parameter $\hat p_i$, rather than checking across all predictions? Why? | |
| Sep 12, 2023 at 23:11 | comment | added | kjetil b halvorsen♦ | I guess I mean that the binomial distribution is developed using theory based on the axioms, so how could it not be consistent with the axioms? | |
| Sep 12, 2023 at 22:10 | comment | added | Dave | For the second point, is all you mean that $p,1-p\in[0,1]$, $p +(1-p)=1$, and $P(\{0\}\cup\{1\})=1=p+(1-p)?$ | |
| Sep 12, 2023 at 21:35 | history | answered | kjetil b halvorsen♦ | CC BY-SA 4.0 |