übergeordnetes Maß

15:12 May 26, 2012 

Lovely Walmart link. I'll sell you one cheaper, promise :-)

14:40 May 26, 2012 

Thanks, Donald. The most useful reference is http://en.wikipedia.org/wiki/Portmanteau_test -- which unfortunately I didn't come up with before, because I used "portmanteau measure" for my search terms! (But in your search link, the one I liked best was from Walmart.com: "Shop for Portmanteau test at Walmart.com and save." Don't you just love the Internet?)

14:26 May 26, 2012 

... ctd.

So, a portmanteau test is any test performed against a portmanteau measure/statistic (this being some measure of a number of different aspects of the data. One tests the value of the statistic (derived from the data) against a value in some lookup table and then decides whether this statistic is "significant" (really means something vs might just have happened in some random way). Fortunately, there's no single lookup table for the "attractiveness statistic" :-)
<br><br><br>
Note is that I'm suggesting "a portmanteau measure", not "the portmanteau measure" (the difference is important here).

The following search might help?
https://www.google.ie/search?num=100&hl=en&lr=&safe=off&as_q...

14:04 May 26, 2012 

There's a reliable answer at
https://stat.ethz.ch/pipermail/r-help/2008-November/179334.h...
"Can somebody please explain me what is Portmanteau test? Why it's name is like that? When I would say, a particular test is portmanteau test? I did some googling but got no satisfactory answer at all. Please anybody help for understanding that?

The ``Portmanteau test'' is the name often given to the Box-Pierce or modified
Box-Pierce (Box-Ljung or Ljung-Box-Pierce) statistic. See e.g. Jonathan Cryer, ``Time Series Analysis'', Duxbury 1986, page 153.

It is called the Portmanteau test because it packs a bunch of tests into a single ``suitcase''. (``Portmanteau'' is an old-fashioned word for suitcase.)"

Think e.g. of deciding if someone is "attractive" - there will be bunch of sub-tests "regular features", "nice smile", "sense of humour". All bundled up into one single decision "attractive?" YES|NO. This single decision is using a portmanteau test on an "attractiveness measure/statistic. ctd later ...

12:47 May 26, 2012 

Your answers seem the most likely and authoritative, but I can't confirm either of them. Neither comes up in searches of EN statistical glossaries, and I can't really figure out what either of them means, apart from the ordinary meanings of the component words of the phrase. And I can't find anything that links them to the German phrase in question. In case it is not obvious, I am not at all knowledgeable about statistics. Is there anything you (or any of the other answerers) can say to convince me that your answer is not just plausible, but also correct? Sorry to be a pain.