Problem 2049. Six Steps to PCA - Step 2: Covariance

Created by Neil Caithness

Solve Later

{"relationships":[{"relationshipType":"http://schemas.mathworks.com/matlab/code/2013/relationships/document","targetMode":"","relationshipId":"rId1","target":"/matlab/document.xml"},{"relationshipType":"http://schemas.mathworks.com/matlab/code/2013/relationships/output","targetMode":"","relationshipId":"rId2","target":"/matlab/output.xml"}],"parts":[{"partUri":"/matlab/document.xml","relationship":[],"contentType":"application/vnd.mathworks.matlab.code.document+xml","content":"<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<w:document xmlns:w=\"http://schemas.openxmlformats.org/wordprocessingml/2006/main\"><w:body><w:p><w:pPr><w:pStyle w:val=\"text\"/></w:pPr><w:r><w:rPr><w:b/></w:rPr><w:t>Previous problems in this series</w:t></w:r></w:p><w:p><w:pPr><w:pStyle w:val=\"ListParagraph\"/><w:numPr><w:numId w:val=\"1\"/></w:numPr></w:pPr><w:hyperlink w:docLocation=\"http://www.mathworks.co.uk/matlabcentral/cody/problems/2043-six-steps-to-pca-step-1-centre-and-standardize\"><w:r><w:t>Step 1: Centre and Standardize</w:t></w:r></w:hyperlink></w:p><w:p><w:pPr><w:pStyle w:val=\"text\"/></w:pPr><w:r><w:rPr><w:b/></w:rPr><w:t>Step 2: Covariance</w:t></w:r></w:p><w:p><w:pPr><w:pStyle w:val=\"text\"/></w:pPr><w:r><w:t>For the second step we want the covariance matrix of the centred and standardized data that we calculated in Step 1. We could conveniently use Matlab's</w:t></w:r><w:r><w:t> </w:t></w:r><w:r><w:rPr><w:rFonts w:cs=\"monospace\"/></w:rPr><w:t>cov</w:t></w:r><w:r><w:t> function, but this repeats much of what we've already done, for instance centering by removing the mean. It also has to deal with a wider range of possible inputs, whereas we have a very well specified starting point.</w:t></w:r></w:p><w:p><w:pPr><w:pStyle w:val=\"text\"/></w:pPr><w:r><w:rPr><w:b/></w:rPr><w:t>Task</w:t></w:r></w:p><w:p><w:pPr><w:pStyle w:val=\"text\"/></w:pPr><w:r><w:t>Build on the code produced for Step 1 by writing a function to calculate the covariance matrix of the centred and standardized data matrix.</w:t></w:r></w:p><w:p><w:pPr><w:pStyle w:val=\"text\"/></w:pPr><w:r><w:t>Add the following field to those already produced by your</w:t></w:r><w:r><w:t> </w:t></w:r><w:r><w:rPr><w:rFonts w:cs=\"monospace\"/></w:rPr><w:t>zscore</w:t></w:r><w:r><w:t> function.</w:t></w:r></w:p><w:p><w:pPr><w:pStyle w:val=\"ListParagraph\"/><w:numPr><w:numId w:val=\"1\"/></w:numPr></w:pPr><w:r><w:rPr><w:rFonts w:cs=\"monospace\"/></w:rPr><w:t>Cov</w:t></w:r><w:r><w:t>: a square matrix of covariances</w:t></w:r></w:p><w:p><w:pPr><w:pStyle w:val=\"text\"/></w:pPr><w:r><w:rPr><w:b/></w:rPr><w:t>Tips</w:t></w:r></w:p><w:p><w:pPr><w:pStyle w:val=\"ListParagraph\"/><w:numPr><w:numId w:val=\"1\"/></w:numPr></w:pPr><w:r><w:t>It's not as complicated as it sounds. Write one new line of code.</w:t></w:r></w:p><w:p><w:pPr><w:pStyle w:val=\"ListParagraph\"/><w:numPr><w:numId w:val=\"1\"/></w:numPr></w:pPr><w:r><w:t>Don't just call Matlab's own</w:t></w:r><w:r><w:t> </w:t></w:r><w:r><w:rPr><w:rFonts w:cs=\"monospace\"/></w:rPr><w:t>cov</w:t></w:r><w:r><w:t> function.</w:t></w:r></w:p><w:p><w:pPr><w:pStyle w:val=\"ListParagraph\"/><w:numPr><w:numId w:val=\"1\"/></w:numPr></w:pPr><w:r><w:t>Keep to the structure of the code template with</w:t></w:r><w:r><w:t> </w:t></w:r><w:r><w:rPr><w:rFonts w:cs=\"monospace\"/></w:rPr><w:t>zscore</w:t></w:r><w:r><w:t> as a local function.</w:t></w:r></w:p></w:body></w:document>"},{"partUri":"/matlab/output.xml","contentType":"text/xml","content":"<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\" ?><embeddedOutputs><metaData><evaluationState>manual</evaluationState><layoutState>code</layoutState><outputStatus>ready</outputStatus></metaData><outputArray type=\"array\"/><regionArray type=\"array\"/></embeddedOutputs>"}]}

Previous problems in this series<ul><li><a href = "http://www.mathworks.co.uk/matlabcentral/cody/problems/2043-six-steps-to-pca-step-1-centre-and-standardize">Step 1: Centre and Standardize</a></li></ul>Step 2: CovarianceFor the second step we want the covariance matrix of the centred and standardized data that we calculated in Step 1. We could conveniently use Matlab's <tt>cov</tt> function, but this repeats much of what we've already done, for instance centering by removing the mean. It also has to deal with a wider range of possible inputs, whereas we have a very well specified starting point.TaskBuild on the code produced for Step 1 by writing a function to calculate the covariance matrix of the centred and standardized data matrix.Add the following field to those already produced by your <tt>zscore</tt> function.<ul><li><tt>Cov</tt>: a square matrix of covariances</li></ul>Tips<ul><li>It's not as complicated as it sounds. Write one new line of code.</li><li>Don't just call Matlab's own <tt>cov</tt> function.</li><li>Keep to the structure of the code template with <tt>zscore</tt> as a local function.</li></ul>

*Previous problems in this series*

* <http://www.mathworks.co.uk/matlabcentral/cody/problems/2043-six-steps-to-pca-step-1-centre-and-standardize Step 1: Centre and Standardize>

*Step 2: Covariance*

For the second step we want the covariance matrix of the centred and standardized data that we calculated in Step 1. We could conveniently use Matlab's |cov| function, but this repeats much of what we've already done, for instance centering by removing the mean. It also has to deal with a wider range of possible inputs, whereas we have a very well specified starting point.

*Task*

Build on the code produced for Step 1 by writing a function to calculate the covariance matrix of the centred and standardized data matrix.

Add the following field to those already produced by your |zscore| function.

* |Cov|: a square matrix of covariances

*Tips*

* It's not as complicated as it sounds. Write one new line of code.
* Don't just call Matlab's own |cov| function.
* Keep to the structure of the code template with |zscore| as a local function.

Solve

Solution Stats

103 Solutions
10 Solvers

Last Solution submitted on Apr 27, 2022

Last 200 Solutions

Problem Comments

2 Comments

2 Comments

Jean-Marie Sainthillier on 10 Feb 2014

Where are the other steps Neil ?

Rafael S.T. Vieira on 2 Sep 2020

The problem wants the covariance matrix of the z-score obtained at the previous step if someone else found the problem description unclear.

Solution Comments

Solution 372377

2 Comments

2 Comments

Neil Caithness on 19 Dec 2013

Now what was so hard about that? :-)
Next I need to work on this pernicious 'ans' rubbish.

J.R.! Menzinger on 19 Dec 2013

HAHAHA I had some fun here, sorry about that... ;-)

but also I forgot the minus one. :-/ So the test-case was wrong. Now I found my mistake.

Solution 372370

1 Comment

1 Comment

J.R.! Menzinger on 19 Dec 2013

:-P

Solution 371678

1 Comment

1 Comment

J.R.! Menzinger on 20 Dec 2013

I had the same idea, but I was occupied by bypassing the covariance function :-D

Problem Recent Solvers10

Problem Tags

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Problem 2049. Six Steps to PCA - Step 2: Covariance

Solution Stats

Last 200 Solutions

Problem Comments

Solution Comments

Problem Recent Solvers10

Suggested Problems

More from this Author1

Problem Tags

Community Treasure Hunt

Problem 2049. Six Steps to PCA - Step 2: Covariance

Solution Stats

Last 200 Solutions

Problem Comments

Solution Comments

Problem Recent Solvers10

Suggested Problems

More from this Author1

Problem Tags

Community Treasure Hunt

Players