Variance-covariance matrix

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

Variance-covariance matrix

Giorgio Garziano
Hi,

I am looking for a R package providing with variance-covariance matrix computation of univariate time series.

Please, any suggestions ?

Regards,

Giorgio


        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
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.
Reply | Threaded
Open this post in threaded view
|

Re: Variance-covariance matrix

David Winsemius

On May 10, 2015, at 4:27 AM, Giorgio Garziano wrote:

> Hi,
>
> I am looking for a R package providing with variance-covariance matrix computation of univariate time series.
>
> Please, any suggestions ?

If you mean the auto-correlation function, then the stats package (loaded by default at startup) has facilities:

?acf
# also same help page describes partial auto-correlation function
#Auto- and Cross- Covariance and -Correlation Function Estimation

--

David Winsemius
Alameda, CA, USA

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
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.
Reply | Threaded
Open this post in threaded view
|

Re: Variance-covariance matrix

Giorgio Garziano
Hi,

Actually as variance-covariance matrix I mean:

        http://stattrek.com/matrix-algebra/covariance-matrix.aspx

that I compute by:

        data <- rnorm(10,2,1)
        n <- length(data)
        data.center <- scale(data, center=TRUE, scale=FALSE)
        var.cov.mat <- (1/(n-1)) * data.center %*% t(data.center)

--
Giorgio Garziano


-----Original Message-----
From: David Winsemius [mailto:[hidden email]]
Sent: domenica 10 maggio 2015 21:27
To: Giorgio Garziano
Cc: [hidden email]
Subject: Re: [R] Variance-covariance matrix


On May 10, 2015, at 4:27 AM, Giorgio Garziano wrote:

> Hi,
>
> I am looking for a R package providing with variance-covariance matrix computation of univariate time series.
>
> Please, any suggestions ?

If you mean the auto-correlation function, then the stats package (loaded by default at startup) has facilities:

?acf
# also same help page describes partial auto-correlation function
#Auto- and Cross- Covariance and -Correlation Function Estimation

--

David Winsemius
Alameda, CA, USA

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
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.
Reply | Threaded
Open this post in threaded view
|

Re: Variance-covariance matrix

Tsjerk Wassenaar
Hi Giorgio,

For a univariate time series? Seriously?

data <- rnorm(10,2,1)
as.matrix(var(data))

Cheers,

Tsjerk


On Sun, May 10, 2015 at 9:54 PM, Giorgio Garziano <
[hidden email]> wrote:

> Hi,
>
> Actually as variance-covariance matrix I mean:
>
>         http://stattrek.com/matrix-algebra/covariance-matrix.aspx
>
> that I compute by:
>
>         data <- rnorm(10,2,1)
>         n <- length(data)
>         data.center <- scale(data, center=TRUE, scale=FALSE)
>         var.cov.mat <- (1/(n-1)) * data.center %*% t(data.center)
>
> --
> Giorgio Garziano
>
>
> -----Original Message-----
> From: David Winsemius [mailto:[hidden email]]
> Sent: domenica 10 maggio 2015 21:27
> To: Giorgio Garziano
> Cc: [hidden email]
> Subject: Re: [R] Variance-covariance matrix
>
>
> On May 10, 2015, at 4:27 AM, Giorgio Garziano wrote:
>
> > Hi,
> >
> > I am looking for a R package providing with variance-covariance matrix
> computation of univariate time series.
> >
> > Please, any suggestions ?
>
> If you mean the auto-correlation function, then the stats package (loaded
> by default at startup) has facilities:
>
> ?acf
> # also same help page describes partial auto-correlation function
> #Auto- and Cross- Covariance and -Correlation Function Estimation
>
> --
>
> David Winsemius
> Alameda, CA, USA
>
> ______________________________________________
> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> 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.
>



--
Tsjerk A. Wassenaar, Ph.D.

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
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.
Reply | Threaded
Open this post in threaded view
|

Re: Variance-covariance matrix

Giorgio Garziano
Hi Tsjerk,

Yes, seriously.

Time series:

X = [x1, x2, x3, ....,xn]

The variance-covariance matrix is V matrix:

            V    =


Σ x12 / (N-1)

Σ x1 x2 / (N-1)

. . .

Σ x1 xn / (N-1)

Σ x2 x1 / (N-1)

Σ x22 / (N-1)

. . .

Σ x2 xn / (N-1)

. . .

. . .

. . .

. . .

Σ xn x1 / (N-1)

Σ xn x2 / (N-1)

. . .

Σ xn2 / (N-1)




Reference: “Time series and its applications – with R examples”, Springer,
     $7.8 “Principal Components” pag. 468, 469

Cheers,

Giorgio


From: Tsjerk Wassenaar [mailto:[hidden email]]
Sent: domenica 10 maggio 2015 22:11
To: Giorgio Garziano
Cc: [hidden email]
Subject: Re: [R] Variance-covariance matrix

Hi Giorgio,

For a univariate time series? Seriously?

data <- rnorm(10,2,1)
as.matrix(var(data))

Cheers,

Tsjerk


On Sun, May 10, 2015 at 9:54 PM, Giorgio Garziano <[hidden email]<mailto:[hidden email]>> wrote:
Hi,

Actually as variance-covariance matrix I mean:

        http://stattrek.com/matrix-algebra/covariance-matrix.aspx

that I compute by:

        data <- rnorm(10,2,1)
        n <- length(data)
        data.center <- scale(data, center=TRUE, scale=FALSE)
        var.cov.mat <- (1/(n-1)) * data.center %*% t(data.center)

--
Giorgio Garziano


-----Original Message-----
From: David Winsemius [mailto:[hidden email]<mailto:[hidden email]>]
Sent: domenica 10 maggio 2015 21:27
To: Giorgio Garziano
Cc: [hidden email]<mailto:[hidden email]>
Subject: Re: [R] Variance-covariance matrix


On May 10, 2015, at 4:27 AM, Giorgio Garziano wrote:

> Hi,
>
> I am looking for a R package providing with variance-covariance matrix computation of univariate time series.
>
> Please, any suggestions ?

If you mean the auto-correlation function, then the stats package (loaded by default at startup) has facilities:

?acf
# also same help page describes partial auto-correlation function
#Auto- and Cross- Covariance and -Correlation Function Estimation

--

David Winsemius
Alameda, CA, USA

______________________________________________
[hidden email]<mailto:[hidden email]> mailing list -- To UNSUBSCRIBE and more, see
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.



--
Tsjerk A. Wassenaar, Ph.D.

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
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.
Reply | Threaded
Open this post in threaded view
|

Re: Variance-covariance matrix

Tsjerk Wassenaar
Hi Giorgio,

This is for a multivariate time series. x1 is variable 1 of the observation
vector x, x2, variable 2, etc. If you need x(i) and x(i+1), etc, then
you're looking for the autocovariance/autocorrelation matrix, which is a
quite different thing (and David showed the way). You can easily see that
you don't have N-1 degrees of freedom per entry, because you have fewer
'observations' for larger lag times.

Cheers,

Tsjerk



On Sun, May 10, 2015 at 10:25 PM, Giorgio Garziano <
[hidden email]> wrote:

>  Hi Tsjerk,
>
>
>
> Yes, seriously.
>
>
>
> Time series:
>
>
>
> X = [x1, x2, x3, ....,xn]
>
>
>
> The variance-covariance matrix is V matrix:
>
>
>
> *            V*    =
>
> Σ *x*12 / (N-1)
>
> Σ *x*1 *x*2 / (N-1)
>
> . . .
>
> Σ *x*1 xn / (N-1)
>
> Σ *x*2 *x*1 / (N-1)
>
> Σ *x*22 / (N-1)
>
> . . .
>
> Σ *x*2 *x*n / (N-1)
>
> . . .
>
> . . .
>
> . . .
>
> . . .
>
> Σ *x*n *x*1 / (N-1)
>
> Σ *x*n *x*2 / (N-1)
>
> . . .
>
> Σ *x*n2 / (N-1)
>
>
>
>
>
> Reference: “Time series and its applications – with R examples”,
> Springer,
>
>      $7.8 “Principal Components” pag. 468, 469
>
>
>
> Cheers,
>
>
>
> Giorgio
>
>
>
>
>
> *From:* Tsjerk Wassenaar [mailto:[hidden email]]
> *Sent:* domenica 10 maggio 2015 22:11
>
> *To:* Giorgio Garziano
> *Cc:* [hidden email]
> *Subject:* Re: [R] Variance-covariance matrix
>
>
>
> Hi Giorgio,
>
>
>
> For a univariate time series? Seriously?
>
>
>
> data <- rnorm(10,2,1)
>
> as.matrix(var(data))
>
>
>
> Cheers,
>
>
>
> Tsjerk
>
>
>
>
>
> On Sun, May 10, 2015 at 9:54 PM, Giorgio Garziano <
> [hidden email]> wrote:
>
> Hi,
>
> Actually as variance-covariance matrix I mean:
>
>         http://stattrek.com/matrix-algebra/covariance-matrix.aspx
>
> that I compute by:
>
>         data <- rnorm(10,2,1)
>         n <- length(data)
>         data.center <- scale(data, center=TRUE, scale=FALSE)
>         var.cov.mat <- (1/(n-1)) * data.center %*% t(data.center)
>
> --
> Giorgio Garziano
>
>
>
> -----Original Message-----
> From: David Winsemius [mailto:[hidden email]]
> Sent: domenica 10 maggio 2015 21:27
> To: Giorgio Garziano
> Cc: [hidden email]
> Subject: Re: [R] Variance-covariance matrix
>
>
> On May 10, 2015, at 4:27 AM, Giorgio Garziano wrote:
>
> > Hi,
> >
> > I am looking for a R package providing with variance-covariance matrix
> computation of univariate time series.
> >
> > Please, any suggestions ?
>
> If you mean the auto-correlation function, then the stats package (loaded
> by default at startup) has facilities:
>
> ?acf
> # also same help page describes partial auto-correlation function
> #Auto- and Cross- Covariance and -Correlation Function Estimation
>
> --
>
> David Winsemius
> Alameda, CA, USA
>
> ______________________________________________
> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> 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.
>
>
>
>
>
> --
>
> Tsjerk A. Wassenaar, Ph.D.
>



--
Tsjerk A. Wassenaar, Ph.D.

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
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.
Reply | Threaded
Open this post in threaded view
|

Re: Variance-covariance matrix

Giorgio Garziano
Hi Tsjerk,

Yes, I understand your point. Thanks for drawing my attention on that aspect.

Let me then rephrase my question.

I would need some R package function able to compute the variance-covariance matrix
for multivariate series as defined at:

        http://stattrek.com/matrix-algebra/covariance-matrix.aspx


About what outlined in the book reference I mentioned, I shall open a separate thread
in the case.

Thanks.

---

Giorgio

Genoa, Italy

From: Tsjerk Wassenaar [mailto:[hidden email]]
Sent: domenica 10 maggio 2015 22:31
To: Giorgio Garziano
Cc: [hidden email]
Subject: Re: [R] Variance-covariance matrix

Hi Giorgio,

This is for a multivariate time series. x1 is variable 1 of the observation vector x, x2, variable 2, etc. If you need x(i) and x(i+1), etc, then you're looking for the autocovariance/autocorrelation matrix, which is a quite different thing (and David showed the way). You can easily see that you don't have N-1 degrees of freedom per entry, because you have fewer 'observations' for larger lag times.

Cheers,

Tsjerk



On Sun, May 10, 2015 at 10:25 PM, Giorgio Garziano <[hidden email]<mailto:[hidden email]>> wrote:
Hi Tsjerk,

Yes, seriously.

Time series:

X = [x1, x2, x3, ....,xn]

The variance-covariance matrix is V matrix:

            V    =


Σ x12 / (N-1)

Σ x1 x2 / (N-1)

. . .

Σ x1 xn / (N-1)

Σ x2 x1 / (N-1)

Σ x22 / (N-1)

. . .

Σ x2 xn / (N-1)

. . .

. . .

. . .

. . .

Σ xn x1 / (N-1)

Σ xn x2 / (N-1)

. . .

Σ xn2 / (N-1)




Reference: “Time series and its applications – with R examples”, Springer,
     $7.8 “Principal Components” pag. 468, 469

Cheers,

Giorgio


From: Tsjerk Wassenaar [mailto:[hidden email]<mailto:[hidden email]>]
Sent: domenica 10 maggio 2015 22:11

To: Giorgio Garziano
Cc: [hidden email]<mailto:[hidden email]>
Subject: Re: [R] Variance-covariance matrix

Hi Giorgio,

For a univariate time series? Seriously?

data <- rnorm(10,2,1)
as.matrix(var(data))

Cheers,

Tsjerk


On Sun, May 10, 2015 at 9:54 PM, Giorgio Garziano <[hidden email]<mailto:[hidden email]>> wrote:
Hi,

Actually as variance-covariance matrix I mean:

        http://stattrek.com/matrix-algebra/covariance-matrix.aspx

that I compute by:

        data <- rnorm(10,2,1)
        n <- length(data)
        data.center <- scale(data, center=TRUE, scale=FALSE)
        var.cov.mat <- (1/(n-1)) * data.center %*% t(data.center)

--
Giorgio Garziano


-----Original Message-----
From: David Winsemius [mailto:[hidden email]<mailto:[hidden email]>]
Sent: domenica 10 maggio 2015 21:27
To: Giorgio Garziano
Cc: [hidden email]<mailto:[hidden email]>
Subject: Re: [R] Variance-covariance matrix


On May 10, 2015, at 4:27 AM, Giorgio Garziano wrote:

> Hi,
>
> I am looking for a R package providing with variance-covariance matrix computation of univariate time series.
>
> Please, any suggestions ?

If you mean the auto-correlation function, then the stats package (loaded by default at startup) has facilities:

?acf
# also same help page describes partial auto-correlation function
#Auto- and Cross- Covariance and -Correlation Function Estimation

--

David Winsemius
Alameda, CA, USA

______________________________________________
[hidden email]<mailto:[hidden email]> mailing list -- To UNSUBSCRIBE and more, see
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.



--
Tsjerk A. Wassenaar, Ph.D.



--
Tsjerk A. Wassenaar, Ph.D.

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
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.
Reply | Threaded
Open this post in threaded view
|

Re: Variance-covariance matrix

Pascal Oettli-2
Hi Giorgio,

No need for a package. Please check function var (?var).

Regards,
Pascal


On Mon, May 11, 2015 at 3:17 PM, Giorgio Garziano
<[hidden email]> wrote:

> Hi Tsjerk,
>
> Yes, I understand your point. Thanks for drawing my attention on that aspect.
>
> Let me then rephrase my question.
>
> I would need some R package function able to compute the variance-covariance matrix
> for multivariate series as defined at:
>
>         http://stattrek.com/matrix-algebra/covariance-matrix.aspx
>
>
> About what outlined in the book reference I mentioned, I shall open a separate thread
> in the case.
>
> Thanks.
>
> ---
>
> Giorgio
>
> Genoa, Italy
>
> From: Tsjerk Wassenaar [mailto:[hidden email]]
> Sent: domenica 10 maggio 2015 22:31
> To: Giorgio Garziano
> Cc: [hidden email]
> Subject: Re: [R] Variance-covariance matrix
>
> Hi Giorgio,
>
> This is for a multivariate time series. x1 is variable 1 of the observation vector x, x2, variable 2, etc. If you need x(i) and x(i+1), etc, then you're looking for the autocovariance/autocorrelation matrix, which is a quite different thing (and David showed the way). You can easily see that you don't have N-1 degrees of freedom per entry, because you have fewer 'observations' for larger lag times.
>
> Cheers,
>
> Tsjerk
>
>
>
> On Sun, May 10, 2015 at 10:25 PM, Giorgio Garziano <[hidden email]<mailto:[hidden email]>> wrote:
> Hi Tsjerk,
>
> Yes, seriously.
>
> Time series:
>
> X = [x1, x2, x3, ....,xn]
>
> The variance-covariance matrix is V matrix:
>
>             V    =
>
>
> Σ x12 / (N-1)
>
> Σ x1 x2 / (N-1)
>
> . . .
>
> Σ x1 xn / (N-1)
>
> Σ x2 x1 / (N-1)
>
> Σ x22 / (N-1)
>
> . . .
>
> Σ x2 xn / (N-1)
>
> . . .
>
> . . .
>
> . . .
>
> . . .
>
> Σ xn x1 / (N-1)
>
> Σ xn x2 / (N-1)
>
> . . .
>
> Σ xn2 / (N-1)
>
>
>
>
> Reference: “Time series and its applications – with R examples”, Springer,
>      $7.8 “Principal Components” pag. 468, 469
>
> Cheers,
>
> Giorgio
>
>
> From: Tsjerk Wassenaar [mailto:[hidden email]<mailto:[hidden email]>]
> Sent: domenica 10 maggio 2015 22:11
>
> To: Giorgio Garziano
> Cc: [hidden email]<mailto:[hidden email]>
> Subject: Re: [R] Variance-covariance matrix
>
> Hi Giorgio,
>
> For a univariate time series? Seriously?
>
> data <- rnorm(10,2,1)
> as.matrix(var(data))
>
> Cheers,
>
> Tsjerk
>
>
> On Sun, May 10, 2015 at 9:54 PM, Giorgio Garziano <[hidden email]<mailto:[hidden email]>> wrote:
> Hi,
>
> Actually as variance-covariance matrix I mean:
>
>         http://stattrek.com/matrix-algebra/covariance-matrix.aspx
>
> that I compute by:
>
>         data <- rnorm(10,2,1)
>         n <- length(data)
>         data.center <- scale(data, center=TRUE, scale=FALSE)
>         var.cov.mat <- (1/(n-1)) * data.center %*% t(data.center)
>
> --
> Giorgio Garziano
>
>
> -----Original Message-----
> From: David Winsemius [mailto:[hidden email]<mailto:[hidden email]>]
> Sent: domenica 10 maggio 2015 21:27
> To: Giorgio Garziano
> Cc: [hidden email]<mailto:[hidden email]>
> Subject: Re: [R] Variance-covariance matrix
>
>
> On May 10, 2015, at 4:27 AM, Giorgio Garziano wrote:
>
>> Hi,
>>
>> I am looking for a R package providing with variance-covariance matrix computation of univariate time series.
>>
>> Please, any suggestions ?
>
> If you mean the auto-correlation function, then the stats package (loaded by default at startup) has facilities:
>
> ?acf
> # also same help page describes partial auto-correlation function
> #Auto- and Cross- Covariance and -Correlation Function Estimation
>
> --
>
> David Winsemius
> Alameda, CA, USA
>
> ______________________________________________
> [hidden email]<mailto:[hidden email]> mailing list -- To UNSUBSCRIBE and more, see
> 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.
>
>
>
> --
> Tsjerk A. Wassenaar, Ph.D.
>
>
>
> --
> Tsjerk A. Wassenaar, Ph.D.
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] mailing list -- To UNSUBSCRIBE and more, see
> 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.



--
Pascal Oettli
Project Scientist
JAMSTEC
Yokohama, Japan

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
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.