"?IC gives you an objective criterion for your model choice, except that it is known to give the wrong answer." Substitute your own favourite letter for ?.
:) Genuine question. There's a perception that we're a bit rough on new users - I don't think that's true - but a Meta thread that appeared to be there just to mock bad questions might give the wrong impression.
On the other hand, with a bit of explanation below each point, it could be useful.
Well, I think we do have a Meta thread on how to ask a good question, and a quite well-upvoted one. Given that, a humorous thread on how to ask a horrible question would I think be perfectly fine. It can actually be quite useful.
Of course. What do you think of the idea of explaining why it's bad below each one? As people do ask questions like these it can't be quite obvious. (Would seem to make it less funny though.)
The likelihood of your sample writes as
$$\begin{align*}
\prod_{i=1}^3 \{p\mathbb{I}_1(x_i)+(1-p) \mathbb{I}_{-1}(x_i)\}&=\prod_{i=1}^3 p^\frac{1+x_i}{2}(1-p)^\frac{1-x_i}{2}\\&=[p(1-p)]^{3/2}
p^\frac{\sum_{i=1}^3 x_i}{2}(1-p)^\frac{-\sum_{i=1}^3x_i}{2}\\
&=[p(1-p)]^{3/2} \sqrt{p/(1-p)}^{\sum_{i=...
I wanted to add one to the list, but I'm not confident enough to take the 1) spot!
2 - epsilon) "Im comparing an XYZ model to a YYZ model using the ABC and DAR statistics. Using the LLV test on the OOP dataset, my results are not good. Perhaps I failed at TTR?
3
I call it the abbreviation stampede.
It hurts me, but seems not to bother others as much, hence my not wanting to take the 1 spot.