Re: [R] How to find more about Covariance (in R)

2011-01-19 Thread Fabrice Tourre
and i.e pearson, kendal and > spearman method) > > > I would like to thank you in advance for your patience. > > Best Regards > Alex > > --- On Wed, 1/19/11, Joshua Wiley wrote: > >> From: Joshua Wiley >> Subject: Re: [R] How to find more about Covariance (in

Re: [R] How to find more about Covariance (in R)

2011-01-19 Thread Steve Lianoglou
Hi, On Wed, Jan 19, 2011 at 9:59 AM, Alaios wrote: > Dear Josh, > I would like to thank you for your reply. > > > I think that it is clear that I miss a lot of theory. I have tried in google > to read and study more about covariance but I have this feeling that the term > 'covariance' is used f

Re: [R] How to find more about Covariance (in R)

2011-01-19 Thread Alaios
hat do not understand i.e pearson, kendal and spearman method) I would like to thank you in advance for your patience. Best Regards Alex --- On Wed, 1/19/11, Joshua Wiley wrote: > From: Joshua Wiley > Subject: Re: [R] How to find more about Covariance (in R) > To: "Alaios" &

Re: [R] How to find more about Covariance (in R)

2011-01-19 Thread Joshua Wiley
Hi Alex, cov() uses the formula for a sample covariance. The denominator is N - 1 instead of N. However, the formula you used from Wikipedia is for a population covariance. To move between the two in this case: cov(sr, sr) * (4/5) # should equal E[XX] - E[x] * E[x] Cheers, Josh On Wed, Jan

[R] How to find more about Covariance (in R)

2011-01-19 Thread Alaios
Hello everyone, I am trying to understand how covariance work. So I created a vector called sr<-c(2,5,7,5,2) so according to wikipedia Cov(X,X)=E[XX]-E[x]*E[x] which in R is mean(sr*sr)-mean(sr)*mean(sr) [1] 3.76 but also cov(sr,sr) [1] 4.7 why is this difference between these two approaches