Apr
7
comment How to deal with subjects who cannot receive the treatment in an experiment?
one more question: What is the difference between 1) and throwing out the untreatable cases?
Apr
5
comment How to deal with subjects who cannot receive the treatment in an experiment?
Thank you very much for the clear and complete answer. It might seem surprising, but we are currently doing 2 and something about it seemed odd to me, hence my question here in SE. Incidentally, a friend mentioned the word "Counterfactual" in this context, but the only thing I could find about it was in the context of observational studies, rather than experiments.
Apr
5
accepted How to deal with subjects who cannot receive the treatment in an experiment?
Apr
3
comment $X_n$ $\,{\buildrel a.s. \over \rightarrow}\,$ $X$, then $(\prod_{i=1}^{n}X_i)^{1/n}$ $\,{\buildrel a.s. \over \rightarrow}\,$ $X$?
$X_i$ has to be strictly positive for this to be meaningful.
Apr
2
comment How to deal with subjects who cannot receive the treatment in an experiment?
Yes, I know at the end of the experiment.
Apr
2
comment How to deal with subjects who cannot receive the treatment in an experiment?
Not in advance. I only get to know only when I want to give them the treatment.
Apr
2
comment How to deal with subjects who cannot receive the treatment in an experiment?
@gung It is a true experiment: the randomization is the first step, as soon as a user comes online.
Apr
2
awarded Popular Question
Apr
2
revised How to deal with subjects who cannot receive the treatment in an experiment?
Added a word to the title to make it better
Apr
2
asked How to deal with subjects who cannot receive the treatment in an experiment?
Mar
26
comment How to Assess the Fit of Thousands of Distributions?
The most practical, and really the best answer, for me is the first comment by @NickCox, but since he didn't put his comment as an answer, and he approves of this answer, I am accepting it as the correct answer.
Mar
26
accepted How to Assess the Fit of Thousands of Distributions?
Mar
24
comment How to Assess the Fit of Thousands of Distributions?
You can't really use KS here, because the parameters for the NULL distribution need to pre-specified in KS. You can generate the NULL distribution using Monte-Carlo, but that's something that I am trying to avoid.
Mar
24
comment How to Assess the Fit of Thousands of Distributions?
@ChrisC Isn't $\chi^2$ for discrete random variables? Here we have a continuous distribution.
Mar
24
comment How to Assess the Fit of Thousands of Distributions?
Ideally, one per subject, but I can live with a single measure.
Mar
24
revised How to Assess the Fit of Thousands of Distributions?
Changed 'test' to 'assess'
Mar
24
comment How to Assess the Fit of Thousands of Distributions?
@Glen_b, Yes, I mean assess. Thanks for pointing it out.
Mar
24
asked How to Assess the Fit of Thousands of Distributions?
Dec
17
awarded Notable Question
Nov
27
comment Can mean plus one standard deviation exceed maximum value?
Maybe she's just thinking of a Normal distribution?
1 2 3 4 5