Statistics: Posted by trackstar — 1 minute ago

]]>

For clarity, I suggest distinguishing the random variable names by 'X' and 'Y' or some such, rather than the -- IMO -- somewhat confusing 'new x value' and 'old x value'.

The general situation is that, if you have a continuous random variable X with density [$]p_X(x)[$], the probability of finding X between 'a' and 'b' is [$]\int_a^b p_X(x) \, dx[$]. If you decide to change variables to [$]Y = f(X)[$], where the mapping is 1-1, then 'a' maps to [$]A = f(a)[$], etc. As you noted, the probability of finding [$]Y[$] between 'A' and 'B' must be exactly the same as the probability of finding X between 'a' and 'b'.

Statistics: Posted by Alan — Yesterday, 8:55 pm

]]>

]]>

Statistics: Posted by Cuchulainn — Yesterday, 7:54 pm

]]>

If my cats were French, I would call them Surrender and Vichy.

Statistics: Posted by katastrofa — Yesterday, 6:37 pm

]]>

Naughty boyAt the risk of undermining the scientific integrity of your process, what result would you like to get?

Statistics: Posted by katastrofa — Yesterday, 4:24 pm

]]>

They are sock suspenders. Garters are worn by ladies and boy scouts.

Point taken. Thank you.and a further thought on your prospective costume. I think it's a winner - something to delight and/or offend everyone.

Lawrence of Arabia.

and make-up inspired by Siouxsie...

Statistics: Posted by trackstar — Yesterday, 4:15 pm

]]>

]]>

@kat I'm not worried about Arabs, I'm worried about

Statistics: Posted by katastrofa — Yesterday, 4:09 pm

]]>

]]>

]]>

Wait! How do you know the color of Paul’s garters?

If search is working, you can find the thread where we were talking about sock garters and he told us he even has a pair in gold lamé.

Not a detail one would forget easily! : )

**

Here you go!

Berlin Code of Conduct - what da?

Statistics: Posted by trackstar — Yesterday, 3:53 pm

]]>

I have two time series and I am testing a NULL hypothesis that volatility of one is large than the vol of the other. For the data, that spans from 1990s to 2017 I get the result - can't reject the NULL at 95% confidence with the p-value (t-stat) very close to the threshold. If I update time series with the 2018 data (now have 1990 -2018), the result flips - NULL can be rejected, and again the p-value (t-stat) very close to the threshold.

Are there other methods, approaches that would produce more stable results ?

Thank you

Statistics: Posted by jgelfand — Yesterday, 3:49 pm

]]>

Sounds fun. Some options:Am invited to an Arabian Nights theme party. What can I wear that isn’t insulting or cultural appropriation? Or should I just drop these friends? Clearly they are not as woke as I.

1) Magician - you even know magic tricks!

2) Sea serpent - blue and green glittery, scaly costume with a long tail.

3) Box of treasure - multicolored silks in patchwork for emeralds, rubies, silver, and gold. And you could wear your gold lame sock garters to match!

I would go as an Arabian horse - and I know how to move like one - all floaty and spirited.

You could too but you would have to practice walk, trot, canter, and spinning turns at home. : )

Will fix this later but here is a picture of an Arabian horse in costume at a show.

My sister has two of them. Beautiful creatures.

Here is the horse - just pick your favorite color scheme and run with it.

Statistics: Posted by bearish — Yesterday, 3:48 pm

]]>

]]>