Winterfors

Thanks for your help with my spelling: I wrote the following on the talk page:

BEGIN: The new opening phrase restricts the article to "linear regression experiments", while the article considers nonlinear experiments and discusses block experiments for categorical variables (ANOVA).

Indeed the literature on block designs for ANOVA exists largely in isolation from the "regression" literature imho (and certainly in the article here!).

For these reasons, I suggest restoring (my?) previous introduction---or writing a new one.

Thank you for your consideration.

Respectfully, END

Would that be agreeable?

BEGIN: Swedish

Hej Winterfors!

Du skriver bättra på engelska än jag kan. Tack för hjälpen med bökstavningerna!

(Is there a spell-check program available here?)

Kull att du är i Paris. På Juisseau, tycker de om Banach spaces, eller?

Mvh, Kiefer.Wolfowitz (talk) 21:02, 24 May 2009 (UTC)

END: Swedish

Hi Kiefer.Wolfowitz,

For the first topic, see the optimal design discussion page.

They do indeed like Banach spaces spaces here, but Sobolev spaces even more at the numerical analysis department I'm at :-) Where are you working yourself? Och hur kommer det sig att du kan svenska?

Fisher information

Hi! I noticed you added Category:Experimental design to the article Fisher information. As far as I can tell the article doesn't have any information on how Fisher information is related to experimental design. Could you elaborate on this? Thanks, --Zvika (talk) 15:47, 27 May 2009 (UTC)

Fisher information matrices are fundamental in optimal experimental design (see optimal design) in applications involving linear models as well as or nonlinear models with approximately Gaussian uncertainties.

I could add something on the topic to the Fisher information article, but the article doesn't contain information on any of the other applications of the Fisher information - so I feel it would be slightly inconsistent to have only one. --Winterfors (talk) 19:03, 27 May 2009 (UTC)

Confirmation of link between Fisher information and optimal designs

Winterfors is correct (as usual)! The Fisher information-matrix is fundamental to the optimal design of experiments (for linear models, etc.).

When writing the paragraphs, I shortened "Fisher information" to "information" to simplify the description, which is already dense and mathematically/statistically demanding. Statisticians already know about Fisher information (one hopes), while non-statisticians may be happy to know that "information" has a precise meaning, for which they may consult the article on (information). Kiefer.Wolfowitz (talk) 19:25, 27 May 2009 (UTC)

The Optimal design article section on variance minimization and Fisher information maximization has been rewritten. I also added (See Also) references to entropy (information theory) and Fisher information. I trust that this addresses the concerns of you (two), and I wish that these changes improve optimal design. Kiefer.Wolfowitz (talk) 22:19, 27 May 2009 (UTC)

I accept the connection, but I think it would be better if an article in the category Category:Experimental design would make some mention of experimental design. Perhaps it would be best to add a section called Applications and within it a subsection on optimal design. It's alright if the section is incomplete - Wikipedia is a work in progress, after all. I might add a section on applications I am familiar with, etc. --Zvika (talk) 05:32, 28 May 2009 (UTC)

Ok, I've added a short word on the application to optimal experimental design to the Fisher information article.

Excellent - that's exactly what I meant. --Zvika (talk) 10:25, 28 May 2009 (UTC)

Your expert feedback is welcome on Template:Experimental design

Hej Emanuel! Det skudde blir trevligt om du skudde titta på Template:Experimental design! Mvh, Kiefer.Wolfowitz (talk) 19:57, 22 June 2009 (UTC)

Invitation to double the size of "Category:Wikipedians interested in statistics"

Hej Emanuel!

It is lonely being the singleton in the "Category:Wikipedians interested in statistics", which now is a supercategory of the "Category:Wikipedians interested in Bayesian methods", whose membership you proudly display!

Please join me!

Mvh, Kiefer.Wolfowitz (talk) 01:15, 14 January 2010 (UTC)

Codomain of a random variable: observation space?

Your opinion is welcome at Wikipedia talk:WikiProject Mathematics#Codomain of a random variable: observation space?. Boris Tsirelson (talk) 16:46, 27 March 2010 (UTC)

Proposed deletion of Mino Raiola

The article Mino Raiola has been proposed for deletion because under Wikipedia policy, all biographies of living persons created after March 18, 2010, must have at least one source that directly supports material in the article.

If you created the article, please don't take offense. Instead, consider improving the article. For help on inserting references, see Wikipedia:Referencing for beginners or ask at Wikipedia:Help desk. Once you have provided at least one reliable source, you may remove the {{prod blp}} tag. Please do not remove the tag unless the article is sourced. If you cannot provide such a source within ten days, the article may be deleted, but you can request that it be undeleted when you are ready to add one. PinkBull 19:52, 3 September 2010 (UTC)