sample covariance matrix











up vote
-1
down vote

favorite












Suppose two covariance function estimators, with the same formula except for a coefficient. Then make two sample covariance matrix(SCM) from each of the functions. Why should these matrices differ in negativeness or positiveness feature? To be clear $R=[R_{i-j}]$ should be non-negative definite(positive semi-definite) but $tilde{R} =[tilde R_{i-j}]$ is not necessarily non-negative definite.
$$y(t)=phi(t)theta$$
$$R_{k}=frac{1}{N}sum_{t=1}^{N-k}y(t)y^T(t+k)quad R_{-k}=R_k^Tquad kge 0$$
$$tilde{R}_{k}=frac{1}{N-k}sum_{t=1}^{N-k}y(t)y^T(t+k)quad tilde R_{-k}=tilde R_k^Tquad kge 0$$
The SCM is built by the following formulla.
$$R=begin{bmatrix}R_0&R_1&cdots&R_{N-k}\
R_1^T&R_0&cdots& R_{N-k-1} \
vdots\
R_{N-k}^T&cdots&&R_0 end{bmatrix}$$

They are used in system identification.
main problem










share|cite|improve this question
























  • What is the matrix here to be clear ? Please make your question non ambiguous, thanks!
    – Ezy
    Nov 17 at 17:07










  • @Ezy Clear enough?
    – V.Ajall
    Nov 18 at 7:14










  • Much better! Now i understand the question :)
    – Ezy
    Nov 18 at 12:15










  • And R is dimension NxN ? In other words you let k run up to N-1 ?
    – Ezy
    Nov 18 at 12:19










  • y is $ntimes 1$ so R is $nNtimes nN$ and yes we let k run up to $N-1$
    – V.Ajall
    Nov 18 at 12:45















up vote
-1
down vote

favorite












Suppose two covariance function estimators, with the same formula except for a coefficient. Then make two sample covariance matrix(SCM) from each of the functions. Why should these matrices differ in negativeness or positiveness feature? To be clear $R=[R_{i-j}]$ should be non-negative definite(positive semi-definite) but $tilde{R} =[tilde R_{i-j}]$ is not necessarily non-negative definite.
$$y(t)=phi(t)theta$$
$$R_{k}=frac{1}{N}sum_{t=1}^{N-k}y(t)y^T(t+k)quad R_{-k}=R_k^Tquad kge 0$$
$$tilde{R}_{k}=frac{1}{N-k}sum_{t=1}^{N-k}y(t)y^T(t+k)quad tilde R_{-k}=tilde R_k^Tquad kge 0$$
The SCM is built by the following formulla.
$$R=begin{bmatrix}R_0&R_1&cdots&R_{N-k}\
R_1^T&R_0&cdots& R_{N-k-1} \
vdots\
R_{N-k}^T&cdots&&R_0 end{bmatrix}$$

They are used in system identification.
main problem










share|cite|improve this question
























  • What is the matrix here to be clear ? Please make your question non ambiguous, thanks!
    – Ezy
    Nov 17 at 17:07










  • @Ezy Clear enough?
    – V.Ajall
    Nov 18 at 7:14










  • Much better! Now i understand the question :)
    – Ezy
    Nov 18 at 12:15










  • And R is dimension NxN ? In other words you let k run up to N-1 ?
    – Ezy
    Nov 18 at 12:19










  • y is $ntimes 1$ so R is $nNtimes nN$ and yes we let k run up to $N-1$
    – V.Ajall
    Nov 18 at 12:45













up vote
-1
down vote

favorite









up vote
-1
down vote

favorite











Suppose two covariance function estimators, with the same formula except for a coefficient. Then make two sample covariance matrix(SCM) from each of the functions. Why should these matrices differ in negativeness or positiveness feature? To be clear $R=[R_{i-j}]$ should be non-negative definite(positive semi-definite) but $tilde{R} =[tilde R_{i-j}]$ is not necessarily non-negative definite.
$$y(t)=phi(t)theta$$
$$R_{k}=frac{1}{N}sum_{t=1}^{N-k}y(t)y^T(t+k)quad R_{-k}=R_k^Tquad kge 0$$
$$tilde{R}_{k}=frac{1}{N-k}sum_{t=1}^{N-k}y(t)y^T(t+k)quad tilde R_{-k}=tilde R_k^Tquad kge 0$$
The SCM is built by the following formulla.
$$R=begin{bmatrix}R_0&R_1&cdots&R_{N-k}\
R_1^T&R_0&cdots& R_{N-k-1} \
vdots\
R_{N-k}^T&cdots&&R_0 end{bmatrix}$$

They are used in system identification.
main problem










share|cite|improve this question















Suppose two covariance function estimators, with the same formula except for a coefficient. Then make two sample covariance matrix(SCM) from each of the functions. Why should these matrices differ in negativeness or positiveness feature? To be clear $R=[R_{i-j}]$ should be non-negative definite(positive semi-definite) but $tilde{R} =[tilde R_{i-j}]$ is not necessarily non-negative definite.
$$y(t)=phi(t)theta$$
$$R_{k}=frac{1}{N}sum_{t=1}^{N-k}y(t)y^T(t+k)quad R_{-k}=R_k^Tquad kge 0$$
$$tilde{R}_{k}=frac{1}{N-k}sum_{t=1}^{N-k}y(t)y^T(t+k)quad tilde R_{-k}=tilde R_k^Tquad kge 0$$
The SCM is built by the following formulla.
$$R=begin{bmatrix}R_0&R_1&cdots&R_{N-k}\
R_1^T&R_0&cdots& R_{N-k-1} \
vdots\
R_{N-k}^T&cdots&&R_0 end{bmatrix}$$

They are used in system identification.
main problem







matrices covariance system-identification






share|cite|improve this question















share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited Nov 18 at 15:18

























asked Nov 17 at 16:36









V.Ajall

74




74












  • What is the matrix here to be clear ? Please make your question non ambiguous, thanks!
    – Ezy
    Nov 17 at 17:07










  • @Ezy Clear enough?
    – V.Ajall
    Nov 18 at 7:14










  • Much better! Now i understand the question :)
    – Ezy
    Nov 18 at 12:15










  • And R is dimension NxN ? In other words you let k run up to N-1 ?
    – Ezy
    Nov 18 at 12:19










  • y is $ntimes 1$ so R is $nNtimes nN$ and yes we let k run up to $N-1$
    – V.Ajall
    Nov 18 at 12:45


















  • What is the matrix here to be clear ? Please make your question non ambiguous, thanks!
    – Ezy
    Nov 17 at 17:07










  • @Ezy Clear enough?
    – V.Ajall
    Nov 18 at 7:14










  • Much better! Now i understand the question :)
    – Ezy
    Nov 18 at 12:15










  • And R is dimension NxN ? In other words you let k run up to N-1 ?
    – Ezy
    Nov 18 at 12:19










  • y is $ntimes 1$ so R is $nNtimes nN$ and yes we let k run up to $N-1$
    – V.Ajall
    Nov 18 at 12:45
















What is the matrix here to be clear ? Please make your question non ambiguous, thanks!
– Ezy
Nov 17 at 17:07




What is the matrix here to be clear ? Please make your question non ambiguous, thanks!
– Ezy
Nov 17 at 17:07












@Ezy Clear enough?
– V.Ajall
Nov 18 at 7:14




@Ezy Clear enough?
– V.Ajall
Nov 18 at 7:14












Much better! Now i understand the question :)
– Ezy
Nov 18 at 12:15




Much better! Now i understand the question :)
– Ezy
Nov 18 at 12:15












And R is dimension NxN ? In other words you let k run up to N-1 ?
– Ezy
Nov 18 at 12:19




And R is dimension NxN ? In other words you let k run up to N-1 ?
– Ezy
Nov 18 at 12:19












y is $ntimes 1$ so R is $nNtimes nN$ and yes we let k run up to $N-1$
– V.Ajall
Nov 18 at 12:45




y is $ntimes 1$ so R is $nNtimes nN$ and yes we let k run up to $N-1$
– V.Ajall
Nov 18 at 12:45










1 Answer
1






active

oldest

votes

















up vote
0
down vote













For $N=2$ and $y(0)=(1,-1)/sqrt{2}$ and $y(1)=(-1,-1)/sqrt{2}$ you see that the first matrix has determinant 0 while the second one has determinant $>0$ so they have different signature.






share|cite|improve this answer























  • please check the changes I have made in the question.
    – V.Ajall
    Nov 18 at 12:46










  • I updated my answer. Going forward you need to make questions non ambiguous
    – Ezy
    Nov 18 at 13:01










  • Thank you but I am looking for the proof of first one not an example only! :)
    – V.Ajall
    Nov 18 at 13:04










  • There is no proof because the claim is wrong. I gave counter example
    – Ezy
    Nov 18 at 13:05










  • Seems that your counter example is wrong since $R_0$ has determinant 0.5 and $R_1$ has determinant 0 and matrix $R$ is nonnegative which is not contradict what has mentioned in the stated problem.
    – V.Ajall
    Nov 18 at 13:34











Your Answer





StackExchange.ifUsing("editor", function () {
return StackExchange.using("mathjaxEditing", function () {
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
});
});
}, "mathjax-editing");

StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "69"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);

StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});

function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
noCode: true, onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});


}
});














draft saved

draft discarded


















StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3002554%2fsample-covariance-matrix%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown

























1 Answer
1






active

oldest

votes








1 Answer
1






active

oldest

votes









active

oldest

votes






active

oldest

votes








up vote
0
down vote













For $N=2$ and $y(0)=(1,-1)/sqrt{2}$ and $y(1)=(-1,-1)/sqrt{2}$ you see that the first matrix has determinant 0 while the second one has determinant $>0$ so they have different signature.






share|cite|improve this answer























  • please check the changes I have made in the question.
    – V.Ajall
    Nov 18 at 12:46










  • I updated my answer. Going forward you need to make questions non ambiguous
    – Ezy
    Nov 18 at 13:01










  • Thank you but I am looking for the proof of first one not an example only! :)
    – V.Ajall
    Nov 18 at 13:04










  • There is no proof because the claim is wrong. I gave counter example
    – Ezy
    Nov 18 at 13:05










  • Seems that your counter example is wrong since $R_0$ has determinant 0.5 and $R_1$ has determinant 0 and matrix $R$ is nonnegative which is not contradict what has mentioned in the stated problem.
    – V.Ajall
    Nov 18 at 13:34















up vote
0
down vote













For $N=2$ and $y(0)=(1,-1)/sqrt{2}$ and $y(1)=(-1,-1)/sqrt{2}$ you see that the first matrix has determinant 0 while the second one has determinant $>0$ so they have different signature.






share|cite|improve this answer























  • please check the changes I have made in the question.
    – V.Ajall
    Nov 18 at 12:46










  • I updated my answer. Going forward you need to make questions non ambiguous
    – Ezy
    Nov 18 at 13:01










  • Thank you but I am looking for the proof of first one not an example only! :)
    – V.Ajall
    Nov 18 at 13:04










  • There is no proof because the claim is wrong. I gave counter example
    – Ezy
    Nov 18 at 13:05










  • Seems that your counter example is wrong since $R_0$ has determinant 0.5 and $R_1$ has determinant 0 and matrix $R$ is nonnegative which is not contradict what has mentioned in the stated problem.
    – V.Ajall
    Nov 18 at 13:34













up vote
0
down vote










up vote
0
down vote









For $N=2$ and $y(0)=(1,-1)/sqrt{2}$ and $y(1)=(-1,-1)/sqrt{2}$ you see that the first matrix has determinant 0 while the second one has determinant $>0$ so they have different signature.






share|cite|improve this answer














For $N=2$ and $y(0)=(1,-1)/sqrt{2}$ and $y(1)=(-1,-1)/sqrt{2}$ you see that the first matrix has determinant 0 while the second one has determinant $>0$ so they have different signature.







share|cite|improve this answer














share|cite|improve this answer



share|cite|improve this answer








edited Nov 18 at 13:01

























answered Nov 18 at 12:23









Ezy

54429




54429












  • please check the changes I have made in the question.
    – V.Ajall
    Nov 18 at 12:46










  • I updated my answer. Going forward you need to make questions non ambiguous
    – Ezy
    Nov 18 at 13:01










  • Thank you but I am looking for the proof of first one not an example only! :)
    – V.Ajall
    Nov 18 at 13:04










  • There is no proof because the claim is wrong. I gave counter example
    – Ezy
    Nov 18 at 13:05










  • Seems that your counter example is wrong since $R_0$ has determinant 0.5 and $R_1$ has determinant 0 and matrix $R$ is nonnegative which is not contradict what has mentioned in the stated problem.
    – V.Ajall
    Nov 18 at 13:34


















  • please check the changes I have made in the question.
    – V.Ajall
    Nov 18 at 12:46










  • I updated my answer. Going forward you need to make questions non ambiguous
    – Ezy
    Nov 18 at 13:01










  • Thank you but I am looking for the proof of first one not an example only! :)
    – V.Ajall
    Nov 18 at 13:04










  • There is no proof because the claim is wrong. I gave counter example
    – Ezy
    Nov 18 at 13:05










  • Seems that your counter example is wrong since $R_0$ has determinant 0.5 and $R_1$ has determinant 0 and matrix $R$ is nonnegative which is not contradict what has mentioned in the stated problem.
    – V.Ajall
    Nov 18 at 13:34
















please check the changes I have made in the question.
– V.Ajall
Nov 18 at 12:46




please check the changes I have made in the question.
– V.Ajall
Nov 18 at 12:46












I updated my answer. Going forward you need to make questions non ambiguous
– Ezy
Nov 18 at 13:01




I updated my answer. Going forward you need to make questions non ambiguous
– Ezy
Nov 18 at 13:01












Thank you but I am looking for the proof of first one not an example only! :)
– V.Ajall
Nov 18 at 13:04




Thank you but I am looking for the proof of first one not an example only! :)
– V.Ajall
Nov 18 at 13:04












There is no proof because the claim is wrong. I gave counter example
– Ezy
Nov 18 at 13:05




There is no proof because the claim is wrong. I gave counter example
– Ezy
Nov 18 at 13:05












Seems that your counter example is wrong since $R_0$ has determinant 0.5 and $R_1$ has determinant 0 and matrix $R$ is nonnegative which is not contradict what has mentioned in the stated problem.
– V.Ajall
Nov 18 at 13:34




Seems that your counter example is wrong since $R_0$ has determinant 0.5 and $R_1$ has determinant 0 and matrix $R$ is nonnegative which is not contradict what has mentioned in the stated problem.
– V.Ajall
Nov 18 at 13:34


















draft saved

draft discarded




















































Thanks for contributing an answer to Mathematics Stack Exchange!


  • Please be sure to answer the question. Provide details and share your research!

But avoid



  • Asking for help, clarification, or responding to other answers.

  • Making statements based on opinion; back them up with references or personal experience.


Use MathJax to format equations. MathJax reference.


To learn more, see our tips on writing great answers.





Some of your past answers have not been well-received, and you're in danger of being blocked from answering.


Please pay close attention to the following guidance:


  • Please be sure to answer the question. Provide details and share your research!

But avoid



  • Asking for help, clarification, or responding to other answers.

  • Making statements based on opinion; back them up with references or personal experience.


To learn more, see our tips on writing great answers.




draft saved


draft discarded














StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3002554%2fsample-covariance-matrix%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown





















































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown

































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown







Popular posts from this blog

AnyDesk - Fatal Program Failure

How to calibrate 16:9 built-in touch-screen to a 4:3 resolution?

QoS: MAC-Priority for clients behind a repeater