Reputation
11,924
Top tag
Next privilege 15,000 Rep.
Protect questions
Badges
18 59
Newest
 Necromancer
Impact
~368k people reached

May
11
awarded  Necromancer
Mar
26
awarded  Constituent
Mar
17
awarded  Caucus
Mar
10
comment is it legitimate to t-test between statistics of a bootsrap?
A t-test is only appropriate for something that has a t distribution, which commonly comes from a situation where an underlying variable has a normal distribution and we're estimating it's mean and variance simultaneously, so the estimated mean has a t distribution. Depending on what f is I don't see why you would expect it to have a t distribution. But I don't think that's the point of your question is it? Perhaps you could clarify?
Feb
26
awarded  Popular Question
Dec
13
awarded  Yearling
Dec
12
awarded  Revival
Nov
6
comment Choosing between LM and GLM for a log-transformed response variable
It's an odd way of putting it I know @whuber but comes from $E(Y) = g^{-1}(X\beta)$ becomes $g(E(Y)) = X\beta$. The point being that the link function goes around the $E(Y)$
Sep
30
awarded  Explainer
Sep
21
awarded  Revival
Sep
12
awarded  Nice Answer
Sep
3
awarded  Enlightened
Sep
3
awarded  Nice Answer
Jul
6
comment Correct terminology for describing relative confidence interval
Thanks; and based on the answer by @tharen and comment by G Jay Kerns, "relative margin of error" is the terminology we now use.
Jul
2
awarded  Curious
Jun
8
comment Is there a name for this chart - sort of a cross between a pie chart and a mekko plot
The end result of the project that started with this question can be seen in the interactive web tool linked to from here: mbie.govt.nz/what-we-do/business-growth-agenda/regions
May
23
awarded  Popular Question
Apr
1
awarded  Nice Answer
Feb
28
awarded  Popular Question
Feb
15
awarded  Revival