Timeline for How can I use PCA to estimate the variance-covariance matrix?
Current License: CC BY-SA 3.0
16 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 20, 2019 at 20:35 | answer | added | user3641187 | timeline score: 0 | |
| Dec 5, 2017 at 7:20 | history | bumped | CommunityBot | This question has answers that may be good or bad; the system has marked it active so that they can be reviewed. | |
| Oct 29, 2017 at 4:53 | answer | added | mgilbert | timeline score: 3 | |
| Oct 14, 2015 at 7:40 | comment | added | Muhammad Husnain | Dear thanks a lot for your consistent help.. Best wishes | |
| Oct 13, 2015 at 21:07 | comment | added | amoeba | Yes, this is the procedure. If your goal is simply to reconstruct covariance matrix using three leading PCs, you don't need to use any rotations at all. | |
| Oct 13, 2015 at 13:00 | comment | added | Muhammad Husnain | And if I am right, I confused about 2nd step of above post i.e. selection of eigenvalues. Whether I have to select eigenvalues before Rotation or after the Rotation-rotation sums of squared loading(after rotation, the relative importance are about optimized). So for calculation VC matrix, which one eigenvalues are more meaningful. Kindly see my this and above post as per the pdf file ( jasonhsu.org/uploads/1/0/0/7/10075125/… ), under estimation of PCA. Kindly guide is it correct way for VC matrix based on PCA. I am very thankful for this. Have best wishes | |
| Oct 13, 2015 at 12:42 | comment | added | Muhammad Husnain | Dear Amoeba, what i now understand, I have 5 returns, run PCA, 1).Find Factor loadings (under component matrix extracted by PCA), I will get a matrix of 5*3(Assume I selected 3 component out of 5 with eigenvalues >1). 2) Find the initial eigenvalues (under total variance explained) of first 3 component, and develop a diagonal matrix of 3*3 having diagonal entries equal to eigenvalues and rest 0. 3) take the transpoe of the matrix acheive by first step. Then multiply three matrix, I will get the 5*5 Variance covariance matrix. kindly confirm am I right based upon the above link in my last post? | |
| Oct 12, 2015 at 10:23 | comment | added | amoeba | I think what is meant there is to compute sample covariance matrix, then do PCA, and keep only few components, i.e. use a low-rank approximation to the sample covariance matrix. Cc to @gung. | |
| Oct 11, 2015 at 16:43 | comment | added | Muhammad Husnain | Dear, thank, here is the link of the pdf file, jasonhsu.org/uploads/1/0/0/7/10075125/… In this file, there are four ways for estimation of VC matricx, On page 3, there is a way for the estimation of VC by PCA. Kindly guide how I can estimate this model. As i know other 3 ways described in that documnet for estimation of covariance matrix. A thousand thanks for this. | |
| Oct 11, 2015 at 12:22 | comment | added | amoeba | Muhammad, please post your image on imgur.com and post a link here in a comment. If it is a pdf file, post in anywhere you want (dropbox? google drive? there are plenty of possibilities) and post a link here too. Alternatively, provide some quotes from this document. Currently your question is unclear and can be closed as such. | |
| Oct 11, 2015 at 9:13 | comment | added | Muhammad Husnain | Dear, I have one page pdf format file, I wana to share it with you/with this post. But I am not able to find the way how to post this image etc. But in that file He describe the ways to estimate the covariance matrix by PCA. and I am not able to understand it. | |
| Oct 10, 2015 at 20:27 | history | edited | amoeba | CC BY-SA 3.0 |
light editing, tags, title
|
| Oct 10, 2015 at 19:52 | history | edited | kjetil b halvorsen♦ | CC BY-SA 3.0 |
edited title
|
| Oct 10, 2015 at 16:22 | comment | added | gung - Reinstate Monica | PCA operates over the covariance matrix. You need to have the covariance matrix first. I don't think this is going to work out. | |
| Oct 10, 2015 at 15:39 | review | First posts | |||
| Oct 10, 2015 at 16:22 | |||||
| Oct 10, 2015 at 15:34 | history | asked | Muhammad Husnain | CC BY-SA 3.0 |