User talk:Orderud

From Wikipedia, the free encyclopedia

Welcome!

Hello, and welcome to Wikipedia. Thank you for your contributions. I hope you like the place and decide to stay. Here are a few good links for newcomers:

I hope you enjoy editing here and being a Wikipedian! By the way, you can sign your name on Talk and vote pages using three tildes, like this: ~~~. Four tildes (~~~~) produces your name and the current date. If you have any questions, see the help pages, add a question to the village pump or ask me on my Talk page. Again, welcome! Oleg Alexandrov 01:56, 21 Apr 2005 (UTC)

Contents

[edit] Kalman filter derivation

There are real problems with your text for the derivation of the Kalman filter, but when I tried to fix them, you reverted all my changes, saying that they were inaccurate. One part I know was inaccurate (and indeed had marked on my printouts to fix before your changes went in): I referred to "estimated state matrix" where what I meant was "estimated state vector", and I didn't make it clear that it was the sum of the squares of the error that was to be minimized.

You reverted all my changes to the original text, so I'm guessing there must have been more serious problems than that. Could you outline them for me? Thanks!

My consern with your rewrite was that you turned the text into an introduction to stochastic estimation (in general) by introducing sentences like we must define precisely what we mean by "best possible estimate". The Kalman article in not the place to describe what MMSE is. However, I agree with you in that the section definetely needs to be rephrased.
What about stating that the method is based on an MMSE solution to the (quadratic) E \{ \textbf{e}_{k|k} \textbf{e}_{k|k}^T \} error metric?
Please don't take this personally. My intension was only to make the article appear more like a research paper introduction to Kalman would have been. --Fredrik Orderud 09:22, 20 Apr 2005 (UTC)
OK, that's very helpful. I've re-worded in accordance with this advice, and also used var to simplify the equations. I hope I haven't done anything that seems disastrous to you! — ciphergoth 11:53, 2005 Apr 20 (UTC)

[edit] Capital letters for random variables

Hello. Standard usage in my experience is that random variables are denoted by capital letters like X, and corresponding lower-case letters used for the dummies in density functions and CDFs and such expressions as

Pr(X > x)
fX(x) = e−x (capital in the subscript, lower-case in parentheses, for obvious reasons)

So why did you change the RVs in covariance matrix to lower-case? Michael Hardy 21:56, 20 Apr 2005 (UTC)

I agree with you when dealing with scalar statistics. However, when dealing with "vector statistics" is is common pracise to denote scalars normal, vectors boldface and matrices capitalized to help distinguish between them. Please respond again if you still disagree, and I'll change the notation back.--Fredrik Orderud 22:12, 20 Apr 2005 (UTC)

[edit] Kalman filter: non-correlation of noise variables

If you have a moment free, I'd appreciate your guidance in a discussion I'm having with User:Chrislloyd on Talk:Kalman filter#non-correlation_of_noise_variables about whether uncorrelation of gaussian variables implies independence. One of us is very confused and wrong, and would benefit from the knowledge of one such as yourself. Thanks! — ciphergoth 20:18, 2005 Apr 27 (UTC)

You can now find my answer in the Kalman talk-page. It's a very interesting discussion you've got :) --Fredrik Orderud 21:37, 27 Apr 2005 (UTC)

[edit] Correlation matrix?

In signal processing, the correlation matrix of a vector x is denoted
\textbf{R}_x = E[ \textbf{x} \textbf{x}^\top ],
and the covariance matrix is denoted
\textbf{C}_x = E[(\textbf{x} - E[\textbf{x}] )(\textbf{\textbf{x}} - E[\textbf{x}] )^\top]
for real signals.

You've got to be kidding. That the difference between covariance and correlation is that in the former you don't subtract the average, is very strange. The standard is that the covariance divided by the product of the two standard deviations is the correlation. Michael Hardy 20:14, 28 Apr 2005 (UTC)

No, I'm not kidding! Just look at p.150 at Discrete Random Signals and Statistical Signal Processing by C. Therrien. This book is described as a "Reference Text" in 6.432 Stochastic Processes, Detection, and Estimation at your own universiy (MIT). This course also uses the same definition for correlation (problem 7.7 b). I fully acknowledge that this definision is inconsistent with the ordinary definition of "correlation", and therefore put it in a "signal processing" section. --Fredrik Orderud 20:51, 28 Apr 2005 (UTC)

[edit] Matrix differentiation

Hi Frederik. I left you a message on Talk:Matrix differentiation. Thanks. Oleg Alexandrov 00:48, 8 May 2005 (UTC)

The article is now renamed to Matrix calculus. --Fredrik Orderud 13:19, 8 May 2005 (UTC)

[edit] Um...

It's clear that that was just vandalism. In cases such as this you just revert back to the last good version. Not sure why you put the stub notice at the top either, as they go at the bottom of the article. CryptoDerk 17:24, May 9, 2005 (UTC)

Did you read the external link that the person provided? The Watley Review is dedicated to the production of articles completely without journalistic merit or factual basis, as this would entail leaving our chairs or actually working. Names, places and events are generally fictitious, except for public figures about which we may have heard something down at the pub. All contents are intended as parody and should be construed as such. When someone adds "facts" to an article and references them with a parody site, I consider that vandalism. CryptoDerk 17:42, May 9, 2005 (UTC)
Sorry. I didn't notice your removal of the links and vandalized content when reverting. I thought you only removed the "disputed" tag. Thanks a lot for you help! --Fredrik Orderud 17:45, 9 May 2005 (UTC)

[edit] Categories for Deletion

  • Hi, I noticed you marked Category:Numerical programming systems for deletion but never added it to WP:CFD. I've removed the tag for now. If you definitely wanted to delete this category, please make sure you complete both parts of the procedure. Thanks! --Kbdank71 17:40, 20 May 2005 (UTC)

[edit] Category:Curves

Hi Orderud. As I think you saw, I reverted some of your edits. I think the numerical analysis category is not so reduntant in places. Also, when one thinks of the Category:Curves, the first thing which comes to one's mind is that this is geometry, not numerical analysis.

Thus, putting Category:Curves in Category:Numerical analysis, then removing the latter from a bunch of articles which are about numerical analysis, I think was not right.

However, in retrospect, I think it was not right to revert your edits so fast (I guess I am in a hurry right now). After I already did that, another idea came to my mind. Category:Curves should not be in Category:Numerical analysis. Rather, a new Category:Splines needs to be created, which should be in the categories of curves, numerical analysis, and computer graphics. Then, the articles about splines which are now in Category:Curves and Category:Numerical analysis could be put in the new category instead. What do you think?

Again, I am sorry for my rush edits. I should have considered this option before. Oleg Alexandrov 04:20, 10 Jun 2005 (UTC)

My intention by recategorization of spline-articles was to reduce the size of the numerical analysis and computer graphics categories, which I think are unneccesarily large. A more precise categorization would reduce this problem.
I totally agree in that a "Splines" category is better for spline-related content than my attemt to use the "Curve" category. I've therefore just created Category:Splines, and moved the first 10 spline related articles to this category. --Fredrik Orderud 10:48, 10 Jun 2005 (UTC)
Hey Fredrik, good work! Thanks! Oleg Alexandrov 15:03, 10 Jun 2005 (UTC)
BTW, are you aware of Wikipedia:WikiProject Mathematics? That, and its talk page are where the mathematicians gather to discuss all math related issues. There is also a list of participants to sign on. Cheers, Oleg Alexandrov 15:03, 10 Jun 2005 (UTC)

[edit] Element densities in infoboxes

We, at the Wikiproject Elements, have noticed your recent edits to the infoboxes of helium and hydrogen. We were wondering your source for the data and if it is reliable. If it is, we would be interesting in expanding it to all emement articles. Please leave us a comment on Wikipedia talk:WikiProject Elements or even my talk page. --metta, The Sunborn 3 July 2005 23:43 (UTC)

[edit] Google is all knowing. All hail mighty Google

Perhaps next time you should consult the oracle... sorry Google. :) AlistairMcMillan 23:21, August 6, 2005 (UTC)

[edit] A Norwegian as well, I presume?

Hi; welcome! I hope it's OK if I (re)add you to the WKPns in Norway cat---I think it's a patently Good Thing™ to have us Norwegians listed in that cat in addition to any 'regional' ones, so as to keep the overall picture. --Wernher 14:25, 21 August 2005 (UTC)

That's ok :) --Fredrik Orderud 14:44, 21 August 2005 (UTC)
I've changed my mind; and do now disagree with you. I think membership in Category:Wikipedians in Norway is unneccesary for members in Category:Wikipedians in Trondheim, as I oppose redundant categorization. --Fredrik Orderud 02:31, 18 September 2005 (UTC)

[edit] Science pearls

Hello,

Since you contributed in the past to the publications’ lists, I thought that you might be interested in this new project. I’ll be glad if you will continue contributing. Thanks,APH 09:39, 11 September 2005 (UTC)

[edit] Video for Windows

About your recent edit...

removing cat:windows software, since this is a library, and not user software

(1) Video for Windows is not just libraries, the actual Media Player application was only distributed as part of Video for Windows. (2) The cat is "Windows software" not "Windows user software". If the cat is only for user software it should say so somewhere. AlistairMcMillan 15:08, 13 September 2005 (UTC)

You might be right there, but isn't Category:Microsoft software still sufficient categorization for Video for Windows? --Fredrik Orderud 15:46, 13 September 2005 (UTC)

[edit] Wikibooks:Programming:MS Visual Studio

PANIC: I've posted a message for you in [Wikibooks user talk page], txs...

[edit] Microsoft Dynamic Link Library → Dynamic Link Library

Please participate in the discussion at Talk:Microsoft Dynamic Link Library#Requested move.

[edit] Please comment on the current Math Collaboration of the Week

Hello Orderud - since you listed statistics as an interest in your user talk, I was hoping you could lend your expertise to the current Mathematics Collaboration of the Week: Multiple Comparisons. Obviously it's a interesting and important topic. We are also in the midst of a discussion as to the distinction between multiple comparisons and multiple testing. Your thoughts would be much appreciated. Let's get a math article up on the front page! Thanks for any help. Debivort 10:20, 5 January 2006 (UTC)

[edit] It's the thought that counts

Yeah, voting closed just yesterday on my RfA, but thanks for the sentiment! -lethe talk + 01:24, 2 February 2006 (UTC)

Thanks for experimenting with Wikipedia. Your test worked, and has been reverted or removed by an automated bot. Please use the sandbox for any other tests you want to do. Take a look at the welcome page if you would like to learn more about contributing to our encyclopedia. Thanks. If you feel you have received this notice in error, please contact the bot owner // Tawkerbot2 20:52, 31 March 2006 (UTC)

[edit] Kalman and HMM

Your revert covers only some of Alan1507's edits. this diff covers the changes between before he started editing and the current version. — ciphergoth 22:44, 11 May 2006 (UTC)

I know. The previous edits by Alan1507 didn't look that bad, so I only reverted the last one where he replaced "Markov chain" with "graphical model". --Fredrik Orderud 22:51, 11 May 2006 (UTC)

[edit] Weather radar

Thanks a lot for the reformating of the equations. I've looked in the Latex Help section but I could not find how to realign the text on the right of the brace. I will keep you changes in my files.

Pierre_cb 13:38 GMT 2006-05-16

Look at m:Help:Formula for a great cheat sheet for latex math :-) --Fredrik Orderud 13:47, 16 May 2006 (UTC)

[edit] Inferring backwards?

A question I asked on Talk:Kalman filter but didn't get a very good answer; I hope you don't mind me asking you directly. A Kalman model will use today's observation to estimate today's state. What do you use when you want to use today's observation (in combination with those from the past) to improve your estimate of yesterday's state? There are definitely well-understood techniques that use both past and future information together; I just don't know how to look for them because I don't know what they're called. Thanks! — ciphergoth 10:53, 11 August 2006 (UTC)

[edit] Weather radar

Hi,

You have put a "Clean up" sign on my Weather radar article in May 2006, before it was completly finished. Since then, I added lot of stuff and reorganized. Some people have added things too. I would like to know want you think of it now and how to get this sign off. Pierre cb 05:19, 27 August 2006 (UTC)

The article sure looks a lot better now, so you can probably remove the "cleanup" tag at the top of the article. --Fredrik Orderud 12:07, 27 August 2006 (UTC)

[edit] Cleanup on Automatically_Tuned_Linear_Algebra_Software

I have been working to cleanup Automatically_Tuned_Linear_Algebra_Software. I see that you have edited that page in the past. If you would be so good as to check out the latest version and contribute to or comment on the article, it would be most appreciated. Cheers, -- Jake 19:36, 5 October 2006 (UTC)

[edit] Image:Visual studio includedirectories.png listed for deletion

An image or media file that you uploaded or altered, Image:Visual studio includedirectories.png, has been listed at Wikipedia:Images and media for deletion. Please look there to see why this is (you may have to search for the title of the image to find its entry), if you are interested in its not being deleted. Thank you.

[edit] Image:Visual studio librarydirectories.png listed for deletion

An image or media file that you uploaded or altered, Image:Visual studio librarydirectories.png, has been listed at Wikipedia:Images and media for deletion. Please look there to see why this is (you may have to search for the title of the image to find its entry), if you are interested in its not being deleted. Thank you.

[edit] Image:Visual studio additionaldependencies.png listed for deletion

An image or media file that you uploaded or altered, Image:Visual studio additionaldependencies.png, has been listed at Wikipedia:Images and media for deletion. Please look there to see why this is (you may have to search for the title of the image to find its entry), if you are interested in its not being deleted. Thank you.

[edit] Image:Visual studio runtimelibrary.png listed for deletion

An image or media file that you uploaded or altered, Image:Visual studio runtimelibrary.png, has been listed at Wikipedia:Images and media for deletion. Please look there to see why this is (you may have to search for the title of the image to find its entry), if you are interested in its not being deleted. Thank you.

[edit] Image tagging for Image:Bland-Altman-Plot.gif

Thanks for uploading Image:Bland-Altman-Plot.gif. The image has been identified as not specifying the source and creator of the image, which is required by Wikipedia's policy on images. If you don't indicate the source and creator of the image on the image's description page, it may be deleted some time in the next seven days. If you have uploaded other images, please verify that you have provided source information for them as well.

For more information on using images, see the following pages:

This is an automated notice by OrphanBot. For assistance on the image use policy, see Wikipedia:Media copyright questions. 10:04, 1 November 2006 (UTC)

Retrieved from "http://en.wikipedia.org../../../o/r/d/User_talk%7EOrderud_bc28.html"