Wikipedia:Reference desk archive/Mathematics/2006 October 10
|< October 9||<<Sep | October | Nov>>||October 11 >|
|The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions at one of the pages linked to above.|
STATA null hypothesis testing
Can someone direct me as to the format to test a null hypothesis in STATA?
I was looking for a pdf for a "crunchy question" (the specific question I was looking at (when I decided to find a formula) was a question which had to do with broken "Crunchy Bars"). When you have n objects, c "correct" objects, taking t objects and need to find the probabilty that you have x of them.
I couldn't find any formula for it, so I devised (?) one.
Is there already a page with such a formula? If not, should I make one? (and what should I call it?)
If I should make a page for this, I suppose I should follow the pattern outlined by the other pages on pdf's (or not)? Is there any page which shows the basic outline required for a page?
One last question regarding "Encyclopedic content must be verifiable" does that mean I need someone to verify it (I already asked my teacher - and he said it looked fine) or does that mean I need sources (what if its original? (I doubt it though))?
- If I understand the question correctly, you are taking a simple random sample without replacement for a dichotomous population. In that case you obtain a hypergeometric distribution. --LambiamTalk 19:29, 10 October 2006 (UTC)
2nd numerability axiom
Can you tell me the statement of the 2nd numerability axiom?
- :P never tought of looking without "axiom", or with "countable", in Mexico is segundo numerable. Thank you
Is there a way, besides guess and check, to find no more than 4 numbers, that when squared and added together, equal 83? (Here's an example: 5= (1)(1) + (2)(2)).
John 17:08, 10 October 2006 (UTC)
3,3,4,7 would be the solution to your question, but then u solve for 5 in a different manner
- To answer the question...no, there isn't any easier way, although I suppose there is a more complex way to do it. StuRat 01:17, 11 October 2006 (UTC)
Yea, I got three different combinations. Just curious. Thanks for your help.John 02:49, 11 October 2006 (UTC)
- I got four: [0, 1, 1, 9]; [0, 3, 5, 7]; [1, 3, 3, 8]; [3, 3, 4, 7]. This was found by an exhaustive search, which is not quite the same as "guess and check". For 8383 there are 202 solutions and not a single one more, which would be hard to assert when using trial and error. --LambiamTalk 05:05, 11 October 2006 (UTC)
Algebra 1 help. I need to find out how to exactly convert percents into degrees so you can put it in a circle graph ans also hot to set up a Relative Freqeuncy Graph
- A full circle (that is: all 100%) is 360°. If you have only 50% (half a circle) you get half of 360° = 180°. 1% means one hundredth of a circle and gives ... . I hope you see the general pattern. --LambiamTalk 20:11, 10 October 2006 (UTC)
- My assumption was that by "a circle graph" the questioner was referring to a pie chart. I'm less sure what kind of visual presentation is referred to by "Relative Freqeuncy Graph". --LambiamTalk 12:11, 12 October 2006 (UTC)
Rays and lines in MS Word
Does anybody know of some sort of downloadable add-on for MS Word that would allow me to put a single or double headed arrow above two letters to indicate the notation for a ray or a line? User101010 19:09, 10 October 2006 (UTC)
- I highly doubt if it's possible to do so in normal text mode, but you can definitely do something like that in Equation mode - go to (IIRC) Tools -> Insert -> Equation (or possibly Tools -> Insert -> Special -> Equation or something of the like), and see if you can work something in that. Confusing Manifestation 08:29, 11 October 2006 (UTC)
- Try Insert → Object..., then from the list choose Microsoft Equation 3.0 and click OK. You'll get a toolbar of equation buttons. Next, click on the button showing a pair of boxes, with a line above the first and another below the second. Then, choose the type of arrow you want. Finally, enter the letters you want to appear below the arrow. --jh51681 22:21, 11 October 2006 (UTC)
Point cloud density
I have am unsorted list of points (xi, yi) with over a million elements. From this list, I want to create a density map. That is, I want to divide the entire domain into a large number of small squares, and color each of the squares according to the number of points contained within that square. Could someone suggest an efficient algorithm to do this? Thanks, deeptrivia (talk) 19:39, 10 October 2006 (UTC)
Okay, I devised one. It was trivial :P
xx = xx-min(xx); yy = yy -min(yy); density=zeros(round(10*max(xx))+1,round(10*max(yy))+1); for i = 1:length(xx) density(round(10*xx(i))+1,round(10*yy(i))+1) = density(round(10*xx(i))+1,round(10*yy(i))+1)+1; end
If the distribution is far from uniform, you might find a quadtree useful.--ColinFine 19:57, 10 October 2006 (UTC)
- I withdraw that - I don't think it will help with this problem. --ColinFine 07:20, 11 October 2006 (UTC)
- The best you can get is O(n) (where n is the number of dots), since at one point or another you have to process each point. Thus, simply looping through the points and adding them to the correct square, is asymptotically optimal. Oskar 19:24, 11 October 2006 (UTC)
Exponential manner of writing
a) how big the distance is expressed by the earth to the 2.25 Mil. light years distant Andromedanebel in km (1 light year = 9,460,000,000,000 km)? Express all figures in exponential manner of writing.
- Welcome to Wikipedia. You can easily look up this topic yourself. Please see Scientific notation. For future questions, try using the search box at the top left of the screen. It's much quicker, and you will probably find a clearer answer. If you still don't understand, add a further question below by clicking the "edit" button to the right of your question title. --ColinFine 19:54, 10 October 2006 (UTC)