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

Fabrice Tourre fabrice.ciup at gmail.com
Wed Jan 19 16:15:06 CET 2011

```population is all individual.
sample is a part of population.
If the sample size is enough large, mean(sr*sr)-mean(sr)*mean(sr) will
almost equal to cov(sr, sr).

On Wed, Jan 19, 2011 at 3:59 PM, Alaios <alaios at yahoo.com> wrote:
> Dear Josh,
>
>
> 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 for many different things in different topics.
>
> You mentioned for example population and sample (covariance) which are two words-terms that I can not distinguish between them.
>
> Even though this is R list could you please provide me some differences to study more about covariance before trying to implement things in R.
>
> (Even in R are some things that 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 <jwiley.psych at gmail.com> wrote:
>
>> From: Joshua Wiley <jwiley.psych at gmail.com>
>> Subject: Re: [R] How to find more about Covariance (in R)
>> To: "Alaios" <alaios at yahoo.com>
>> Cc: r-help at stat.math.ethz.ch
>> Date: Wednesday, January 19, 2011, 2:36 PM
>> 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 19, 2011 at 6:29 AM, Alaios <alaios at yahoo.com>
>> wrote:
>> > 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?
>> Where I am wrong?
>> >
>> > I would like to thank you in advance for your help
>> >
>> > Best Regards
>> > Alex
>> >
>> > ______________________________________________
>> > R-help at r-project.org
>> mailing list
>> > https://stat.ethz.ch/mailman/listinfo/r-help
>> > and provide commented, minimal, self-contained,
>> reproducible code.
>> >
>>
>>
>>
>> --
>> Joshua Wiley
>> Ph.D. Student, Health Psychology
>> University of California, Los Angeles
>> http://www.joshuawiley.com/
>>
>
>
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help