meta user
network profile
| bio | website | |
|---|---|---|
| location | ||
| age | ||
| visits | member for | 9 months |
| seen | 1 hour ago | |
| stats | profile views | 71 |
|
1h |
comment |
estimate population percentage within an interval, given a small sample Yes but I think he wants to take the advantage of the Gaussian distributional assumption. |
|
1h |
comment |
estimate population percentage within an interval, given a small sample I'm under the impression you're on the right way now. You are right: a kind of "inversion" of tolerance intervals provide a confidence interval about a "one-sided" probability $\Pr(X>t)$, but that does'nt work for $\Pr(a \leq X \leq b)$. This problem is straightforward with Bayesian statistics, I hope I'll find the time to write an answer if nobodyelse does it. |
|
1d |
comment |
Symmetry of Plancherel measure (for $S_n$) Thank you (though I will need some time to understand your answer). About by 2nd edit, I am under the impression that for $n \geq 8$ it is possible to have a symmetric bar chart up to one point (i.e. we allow to remove one point). But I need to check again with the computer. |
|
1d |
revised |
Symmetry of Plancherel measure (for $S_n$) added 241 characters in body |
|
1d |
awarded | Tumbleweed |
|
May 18 |
comment |
Hypothesis testing from negative binomial data There is a straightforward conjugate Bayesian analysis to do so. Using the Jeffreys prior it also provides a frequentist solution. |
|
May 18 |
comment |
What is the name of this metric: Why is $(\mathcal{M}, L)$ complete As well as en.wikipedia.org/wiki/Wasserstein_metric and hal.archives-ouvertes.fr/hal-00453887 |
|
May 14 |
revised |
Symmetry of Plancherel measure (for $S_n$) added 7 characters in body |
|
May 14 |
asked | Symmetry of Plancherel measure (for $S_n$) |
|
May 6 |
awarded | Caucus |
|
May 2 |
comment |
Appropriate statistical test to test if probabilities are accurate Cool question. I wonder whether it is more appropriate for stats.stackexchange.com |
|
May 2 |
comment |
Appropriate statistical test to test if probabilities are accurate Are there some "ties" in the first column ? Actually you should describe more precisely how these data are generated. |
|
May 2 |
comment |
Appropriate statistical test to test if probabilities are accurate If there's no link between the tests I'm afraid this is not possible. That would mean that the first line of your data is the outcome of one experiment with "expected" probability $0.09$. The second is the outcome of another experiment, independent of the first one, with expected probability $0.1$. And so on. So your problem is "linewise" (there is an experiment at each line, and the experiments are independent from each other), but at each line you only have one outcome hence you cannot test the hypothesis. |
|
May 2 |
comment |
Appropriate statistical test to test if probabilities are accurate Are the tests independent ? |
|
Apr 28 |
comment |
Since when is 9/10 = 92%? @Shahab Sure but there's no consensual definition. |
|
Apr 28 |
comment |
Since when is 9/10 = 92%? @LukeAshford This is precisely what I was trying to understand. IMHO there's no standard solution to this problem. |
|
Apr 28 |
comment |
Since when is 9/10 = 92%? @Simone I'm not trying to suggest better methods. I'm trying to understand where do the probabilities (92%,...) given by the article come from. That does not sound sensible. |
|
Apr 28 |
comment |
Since when is 9/10 = 92%? @Shabab and you will get $\Pr(p=0.5) =0$ ! The only way to get a positive probability is to use a Bayesian approach with a prior distribution on $p$ having a point mass at $0.5$. |
|
Apr 28 |
comment |
Since when is 9/10 = 92%? @Shahab No, the example you talk about is the probability $\Pr(0.45\leq p \leq 0.55)$. |
|
Apr 28 |
comment |
Since when is 9/10 = 92%? @Shahab I'm statistician. If I well understand the problem, the only way to derive a probability $\Pr(p=0.5)$ that a dice is not biased from the results of dice throwings is a Bayesian-like approach. And there's no standard way to do that. |
