Reputation
3,545
Next privilege 5,000 Rep.
Approve tag wiki edits
Badges
1 16 36
Newest
 Necromancer
Impact
~273k people reached

Jan
28
awarded  Necromancer
Jan
13
comment Is it meaningful to test for normality with a very small sample size (e.g., n = 6)?
@Silverfish He can incorporate that in his answer. The OP clearly stated that he's interested in that normality test in the context of a T-test. In that context, you want to check whether your data doesn't deviate enough from normality to make the T-test lose power. Hence you're interested in the null hypothesis and not the alternative. Hence my focus on the power of these tests. And hence my answer that with N=6, even the most powerful tests can't distinguish between a normal and a heavily skewed distribution. In that context it does not make sense.
Nov
19
comment Binomial confidence interval estimation - why is it not symmetric?
@StephanKolassa You can find the Clopper Pearson formulae here as well : en.wikipedia.org/wiki/…
Nov
4
awarded  Taxonomist
Aug
31
awarded  Yearling
Jul
18
awarded  Popular Question
Jul
14
awarded  Nice Answer
Feb
2
awarded  Pundit
Jan
12
comment Variance on the sum of predicted values from a mixed effect model on a timeseries
@user52220 That's where you're wrong. E(Xi) is the expected value and hence a random variable, whereas mu_i is the mean of the population and hence a fixed number. Var(mu) = 0, but the same is not correct for E(Xi).
Dec
17
awarded  Great Answer
Dec
1
awarded  Enlightened
Dec
1
awarded  Nice Answer
Sep
30
awarded  Explainer
Aug
31
awarded  Yearling
Jul
2
awarded  Curious
May
22
awarded  Popular Question
May
12
awarded  Popular Question
Apr
30
awarded  Notable Question
Feb
5
awarded  Revival
Dec
18
comment Change confidence risk of chisq
Feel free to edit. I did my best, but don't know how to call "random variation" like my grandmother understands it when talking to other statisticians. Randomness is part of the model, as in the random error. But randomness is also part of the experimental setup, as random selection from the population. Hence, a non-random effect in a model can be caused by random variation nonetheless.