cov(A) will compute a covariance matrix of the columns of a matrix A.  cov(a, b) will compute a covariance of the vectors a and b.  here are some examples:  #create random vectors a=rnorm(10) b=rnorm(10)   > a  [1] -0.3110384 -2.6201009 -1.3078118 1.3384743 2.4032381 -0.7245201  [7] -0.2274577 -1.4619624 2.2165118 1.3621709  > b  [1] -1.1648495 -2.3013411 1.5335541 -0.2165172 1.1157774 1.0360323  [7] -0.8309283 -0.3403514 0.3640117 -0.6497384
#create a matrix A=cbind(a,b)  > A               a         b  [1,] -0.3110384 -1.1648495  [2,] -2.6201009 -2.3013411  [3,] -1.3078118 1.5335541  [4,] 1.3384743 -0.2165172  [5,] 2.4032381 1.1157774  [6,] -0.7245201 1.0360323  [7,] -0.2274577 -0.8309283  [8,] -1.4619624 -0.3403514  [9,] 2.2165118 0.3640117 [10,] 1.3621709 -0.6497384  cov(a,b)  Covariance > cov(a,b) [1] 0.7481056  #Covariance matrix cov(A)  > cov(A)          a        b a 2.8529990 0.7481056 b 0.7481056 1.3910519  By the way, the covariance of two scalars will be zero, since the scalars are constants.   I hope this helps.  Sincerely yours,  Mark J. Lamias ________________________________ From: huang jialin <huangps...@gmail.com> To: "Doran, Harold" <hdo...@air.org> Cc: "r-help@r-project.org" <r-help@r-project.org> Sent: Monday, March 12, 2012 12:08 PM Subject: Re: [R] how to calculate a variance and covariance matrix for a vector Doran, Thanks for your advice. Sincerely, Jialin Huang On Mon, Mar 12, 2012 at 10:56 AM, Doran, Harold <hdo...@air.org> wrote: > Always start with the help page. But, the best way to get help on this > list is to construct some toy data. Even if imperfect, people will work to > help you. Maybe show us some of your actual data.**** > > ** ** > > Of course, if a and b1 are really scalars, I donât think there is such a > thing as the covariance between two scalars**** > > ** ** > > ** ** > > * * > > ** ** > > *From:* huang jialin [mailto:huangps...@gmail.com] > *Sent:* Monday, March 12, 2012 11:53 AM > *To:* Doran, Harold > *Cc:* r-help@r-project.org > *Subject:* Re: [R] how to calculate a variance and covariance matrix for > a vector**** > > ** ** > > Hi,**** > > ** ** > > Thanks for your response. **** > > ** ** > > I tried to use cov() before, but got NA all over the place.**** > > ** ** > > a and b1 are scalars in the example. So that is the problem. I will start > over from the formula. **** > > ** ** > > Thanks again.**** > > ** ** > > Sincerely,**** > > Jialin Huang**** > > ** ** > > On Mon, Mar 12, 2012 at 10:47 AM, Doran, Harold <hdo...@air.org> wrote:*** > * > > Can you provide sample data? This suggests that a and b1, for example, are > scalars. If that is what you actually have then you cannot produce a > covariance between two scalars.**** > > > > -----Original Message----- > > From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] > On > > Behalf Of huang jialin > > Sent: Monday, March 12, 2012 11:43 AM > > To: r-help@r-project.org > > Subject: [R] how to calculate a variance and covariance matrix for a > vector > > > > Hello, > > > > I have a vector {a, b1, b2, b3, b4}. How can I calculate the following > > matrix: > > > > var(a) cov(a, b1) cov(a, b2) cov(a, b3) cov(a, b4) > > cov(a, b1) var(b1) cov(a, b2) cov(a, b3) cov(a, b4) > > ... > > ... > > cov(a, b1) cov(a, b2) cov(a, b3) cov(a, b4) var(b4) > > > > I would very appreciate your inputs. Thank you very much. > > > > Sincerely, > > Jialin Huang > >**** > > >   [[alternative HTML version deleted]]**** > > > > > ______________________________________________ > > R-help@r-project.org mailing list > > https://stat.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > > and provide commented, minimal, self-contained, reproducible code.**** > > ** ** >    [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]]
______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
