mm-4169 === Subject: Re: a question about principal component analysis > ..... > I need some help understanding the result of doing PCA on a given dataset. > .... You may like to try sending this question to the news group. Ken Pledger. === Subject: Re: help with math translation from French W. Dale Hall schreef in I ask you to translate the following paper of mathematics (it > consists in four lines) form French to English, please. > http://i7.tinypic.com/4uyknyu.jpg > I beg you to help me, it's very important for me. > Giusy > The original: V_n est un ouvert de M_(n+1). La restriction de phi_n > a V_n est un plongement d'un ouvert de M_(n+1) dans E. > On .8etend cette application en un plongement de M_(n+1) > dans E. En appliquant un th.8eor.8fme d'isotopie des voisinages > tubulaires: on .8etend le plongement en un plongement d'un > voisinage tubulaire de M_(n+1) dans E de sorte que le > diagremme suivant commute; My pretend translation: V_n is an open subset of M_(n+1). The restriction of > phi_n to V_n is an immersion of an open subset of > M_(n+1) into E. We extend this map to an immersion of > M_(n+1) into E. Applying the theorem on isotopy of > tubular neighborhoods: we extend the immersion to an > immersion of a tubular neighborhood of M_(n+1) into E > so that the following diagram commutes: I've modified the literal word-by-word translation to make > the English version read more like it should in how the > terms are used, and I think it's more or less correct. I'm > hoping more, rather than less. Dale plongement -- embedding ? cor === Subject: cauchy problem for I order PDE I am soni manoj from India, when a cauchy problem for I order PDE has unique, NO solution and infinite no of solution? === Subject: Re: RSA & Cryptography > Could anybody explain why RSA works? Why Eucild' Algorithm works? Can > anybody come up with the proof of it? Did you try http://en.wikipedia.org/wiki/RSA http://en.wikipedia.org/wiki/Euclidean_algorithm as a starter? === Subject: NASA Extends Bioastronautics Contract with Wyle Labs 'NASA has exercised a $294 million option to extend a contract with Wyle Laboratories Inc. of Houston to support the Space Life Sciences Directorate at NASA's Johnson Space Center.'Report available at http://theanalystmagazine.com/pr/536.htm === Subject: Re: Solution manual of textbooks in ebook format! Get it in hours! <25072386.1178970034064.JavaMail.jakarta@nitrogen.mathforum.org> Could you email me teh solution manual for Applied Numerical Methods with Matlab for Engineers and Scientists 2nd Ed. by Chapra How much does it cost? === Subject: Re: +1600 Solutions Manual to low cost can i get solutions for Advanced Strength and Applied Elasticity (4th Ed., Ugural) send to same email === Subject: some integration questions hi, i am going over multiple integrals for the test and i came across a few examples which are unclear. it has to do with some really basic rule that i am missing. here's one: /pi/2 / cosx / pi/2 | | e^sinx dr dx - as a result of first integration we end up getting | e^sinx cosx dx = /0 / 0 / 0 | pi/2 = e^sinx| ....my question is how did we end up getting e^sinx after integrating e^sinx cosx? |0 and a very simple integration that i am not following...why do we get - cosx^2/2 when we integrate xsinx^2? thats all, folks === Subject: Re: some integration questions completely screwed up the format..here's the condensed version. = why do we get e^sinx when we integrate e^sinx cos x? = why do we get -cosx^2/2 when we integrate xsinx^2? === Subject: Re: some integration questions completely screwed up the format..here's the condensed version. = why do we get e^sinx when we integrate e^sinx cos x? Why not ask yourself: what is the derivative of e^{sin x}? > = why do we get -cosx^2/2 when we integrate xsinx^2? Why not ask yourself: what is the derivative of -cos(x^2/2)? -- Remove antispam and .invalid for e-mail address. He that giveth to the poor lendeth to the Lord, and shall be repaid, said Mrs Fairchild, hastily slipping a shilling into the poor woman's hand. === Subject: Re: some integration questions >completely screwed up the format..here's the condensed version. = why do we get e^sinx when we integrate e^sinx cos x? Make the substitution u = sin(x) du = cos(x) dx, giving Int e^u du = e^u = e^sin(x) >= why do we get -cosx^2/2 when we integrate xsinx^2? Make the substitution u = x^2 du = 2x dx du/2 = x dx, giving Int 1/2 sin(u) du = -1/2 cos(u) = -1/2 cos(x^2) > === Subject: locus The line y=mx+c meets the lines ax^2+2hxy+by^2 = 0 at A and B and the lines through A and B perpendicular to OA and OB respectively meet at P. Show that, as c varied, the locus of P is the line (a-b)(x-my)+ 2h(mx+y) = 0 === Subject: cartesian plane geometry A triangle with two of its sides on along the lines ax^2 + 2hxy +by^2=0 has its orthocentre at (p,q). Prove that in general, its third side has the equation (a+b)(px+ qy)= aq^2 - 2hpq+ bq^2. Examine the special case when a+b= 0 === Subject: locus what is the locus of the point of intersection of the lines xcos .83+ ysin .83= a , xsin .83 - ycos .83 =a i know it's x^2 + y^2 = 2a^2 but i dont get why this is the locus??? === Subject: Re: locus > what is the locus of the point of intersection of the > lines > xcos .83+ ysin .83= a , xsin .83 - ycos .83 =a i know it's x^2 + y^2 = 2a^2 > but i dont get why this is the locus??? No, you don't know that! You have two equations in two unknowns. You have combined them to make a single equation but haven't eliminated one unknown so you can solve for the other. You have yet another equation that x,y must satisfy: the intersection of the two lines must lie ON that circle, not BE that circle! Multiply the first equation by cos .83: x cos^2 .83+ y sin.83cos.83= a cos.83 Multiply the second equation by sin .83: x sin^2 .83- y sin.83cos.83= a sin.83 Adding those, x= a(cos.83+ sin.83) Multiply the first equation by sin .83: x sin.83cos.83+ y sin^2 .83= a sin .83 Multiply the second equation by -cos.83 -x sin.83cos.83+ y cos^2 .83= -a cos .83 Adding those, y= a(sin.83- cos.83) (x,y)= (a(cos.83+ sin.83),a(sin.83- cos.83)) is a single point that happens to lie on the circle x^2+ y^2= a^2. === Subject: how do i get the mode for a grouped data is there anyone who knows any other formula for getting the mode for a grouped data? i used the formula [mode=Lmo+(d1/(d1+d2))i] but the answer was wrong. perhaps it has something to do with the problem that the highest frequency is located on the topmost class interval and there is no frequency above so that d1 cannot be obtained. what do we do with this type of problem? here's the problem: classes: 85<91, 91<97, 97<103, 103<109, 109<115, 115<121 frequency: 10, 8, 3, 8,0,7(respectively). my answer is 90 but it's incorrect. please help. i want to know the answer badly. if you have anything to say please email me at karryl_041389@yahoo.com === Subject: Re: how do i get the mode for a grouped data > is there anyone who knows any other formula for getting the mode for a grouped data? i used the formula [mode=Lmo+(d1/(d1+d2))i] but the answer was wrong. perhaps it has something to do with the problem that the highest frequency is located on the topmost class interval and there is no frequency above so that d1 cannot be obtained. what do we do with this type of problem? here's the problem: > classes: 85<91, 91<97, 97<103, 103<109, 109<115, 115<121 > frequency: 10, 8, 3, 8,0,7(respectively). my answer is 90 but it's incorrect. please help. i want to know the answer badly. if you have anything to say please email me at karryl_041...@yahoo.com If you have no reason to believe otherwise, then the best you can do is to assume all data points are clustered at the middle of their classes. Since the first class here has the highest frequency, the mode becomes 88 (the middle of the class). Personally, I don't think a grouped mode has much meaning in most cases. (an exception: if the numbers are numerical grades and you want to get a most-common letter grade) The grouped mean is probably more useful in most applications (if you don't have access to the original data, of course), or possibly the median. In the mean case, assume all values are at the middle of their classes. For the second, assume they are evenly distributed through each class. === Subject: Re: Sexual fascism and the Mark Foley scandel On Sun, 24 Jun 2007 06:59:58 -0700, bobandcarole >Apparently, Rep. Mark Foley is homosexual. And like many homosexual > men, he has pedophilic tendencies. We should pray for him that he > gets a handle on this problem and refrains from harming any more > kids. > Meanwhile, we need to wake up. The fact that this is typical > behavior for homosexuals doesn't stop us from continuing to elevate > such folks to positions where they gain access to our kids. > I have many times read that most pedophiles are heterosexuals. http://www.crisisconnectioninc.org/sexualassault/pedophilia_and_child_sexual _mole.htm 95% of known Pedophiles are heterosexual Can you provide a source for saying they are homosexuals?