User talk:Wikiwocow

Your submission at Articles for creation: Metric entropy (October 29)
 Thank you for your recent submission to Articles for Creation. Your article submission has been reviewed. Unfortunately, it has not been accepted at this time. Please view your submission to see the comments left by the reviewer. You are welcome to edit the submission to address the issues raised, and resubmit if you feel they have been resolved.
 * If you would like to continue working on the submission, you can find it at.
 * To edit the submission, click on the "Edit" tab at the top of the window.
 * If you need any assistance, you can ask for help at the [ Articles for creation help desk], or on the [ . Please remember to link to the submission!
 * You can also get live chat help from experienced editors.
 * Thank you for your contributions to Wikipedia! DPRoberts534 (talk) 05:47, 29 October 2013 (UTC)

Your submission at Articles for creation: Metric entropy number (October 29)
 Thank you for your recent submission to Articles for Creation. Your article submission has been reviewed. Unfortunately, it has not been accepted at this time. Please view your submission to see the comments left by the reviewer. You are welcome to edit the submission to address the issues raised, and resubmit if you feel they have been resolved.
 * If you would like to continue working on the submission, you can find it at.
 * To edit the submission, click on the "Edit" tab at the top of the window.
 * If you need any assistance, you can ask for help at the [ Articles for creation help desk], or on the [ . Please remember to link to the submission!
 * You can also get live chat help from experienced editors.
 * Thank you for your contributions to Wikipedia! Rankersbo (talk) 08:44, 29 October 2013 (UTC)

Your submission at Articles for creation: Metric entropy number (October 30)
 Thank you for your recent submission to Articles for Creation. Your article submission has been reviewed. Unfortunately, it has not been accepted at this time. Please view your submission to see the comments left by the reviewer. You are welcome to edit the submission to address the issues raised, and resubmit if you feel they have been resolved.
 * If you would like to continue working on the submission, you can find it at.
 * To edit the submission, click on the "Edit" tab at the top of the window.
 * If you need any assistance, you can ask for help at the [ Articles for creation help desk], or on the [ . Please remember to link to the submission!
 * You can also get live chat help from experienced editors.
 * Thank you for your contributions to Wikipedia! -- t  numbermaniac  c  02:03, 30 October 2013 (UTC)

Your draft article, Wikipedia talk:Articles for creation/Metric entropy number


Hello Wikiwocow. It has been over six months since you last edited your WP:AFC draft article submission, entitled "Metric entropy number".

The page will shortly be deleted. If you plan on editing the page to address the issues raised when it was declined and resubmit it, simply and remove the  or  code. Please note that Articles for Creation is not for indefinite hosting of material deemed unsuitable for the encyclopedia mainspace.

If your submission has already been deleted by the time you get there, and you want to retrieve it, copy this code:, paste it in the edit box at this link , click "Save page", and an administrator will in most cases undelete the submission.

Thanks for your submission to Wikipedia, and happy editing. Rankersbo (talk) 06:00, 30 April 2014 (UTC)

Asymptotic minimax estimators
Hi, I noticed that you added the section "asymptotic minimax estimators" to the article describing "minimax estimators". Rather than editing the article, I thought it's better to talk: I think asymptotic would only be used in a context when some parameter (sample size, confidence parameter) is let to its extreme value. Since there is no parameter in the abstract setting discussed in that section, the article could use some editing. I believe what you define is called near-minimax, or c-approximately minimax (in computing science, it would be called c-competitive, but that's not a terminology that would have been used in statistics). Nevertheless, asymptotic minimax is also important so rather than stripping it, maybe the best way to do is to explain that in some settings there is a parameter etc. By the way, the relaxation of allowing $$c>1$$ normally would not be part of the definition of asymptotic minimax. To describe what I believe is asymptotic minimax in a specific case imagine that $$n$$ is a parameter, and asymptotics is concerned with $$n\to \infty$$ (e.g., $$n$$ could be the sample size). Then $$\delta'$$ asymptotically minimax if for $$\limsup_{n\to\infty} \frac{\sup_{\theta\in \Theta} R_n(\theta,\delta')}{\inf_\delta \sup_\theta R_n(\theta,\delta)}\le 1$$. In these settings, $$\sup_\theta R_n(\theta,\delta')$$ usually converges to zero, so this means that asymptotically $$\sup_\theta R_n(\theta,\delta')$$ converges to zero exactly at the same speed as the minimax risk. If $$1$$ on the right hand side of the definition is replaced by $$c>1$$, we would get the definition of approximately asymptotically minimax. A sidenote is that the above definition requires that $$\delta'$$ is given meaning in the context of all the problems corresponding different values of $$n$$. If you agree, you could edit the article, or I could also do it, if you prefer that. Cheers, --Csaba Szepesvari (talk) 14:44, 10 September 2016 (UTC)