**NOTE: I am currently doing freelance consulting work for machine learning solutions. For all enquires please contact me at george(dot)m(dot)dallas@gmail(dot)com (replace (dot) with a . – this is to prevent spam bots)**

Having been in the social sciences for a couple of weeks it seems like a large amount of quantitative analysis relies on Principal Component Analysis (PCA). This is usually referred to in tandem with eigenvalues, eigenvectors and lots of numbers. So what’s going on? Is this just mathematical jargon to get the non-maths scholars to stop asking questions? Maybe, but it’s also a useful tool to use when you have to look at data. This post will give a very broad overview of PCA, describing eigenvectors and eigenvalues (which you need to know about to understand it) and showing how you can reduce the dimensions of data using PCA. As I said it’s a neat tool to use in information theory, and even though the maths is a bit complicated, you only need to get a broad idea of what’s going on to be able to use it effectively.

There’s quite a bit of stuff to process in this post, but i’ve got rid of as much maths as possible and put in lots of pictures.

## What is Principal Component Analysis?

First of all Principal Component Analysis is a good name. It does what it says on the tin. PCA finds the principal components of data.

It is often useful to measure data in terms of its principal components rather than on a normal x-y axis. So what are principal components then? They’re the underlying structure in the data. They are the directions where there is the most variance, the directions where the data is most spread out. This is easiest to explain by way of example. Here’s some triangles in the shape of an oval:

Imagine that the triangles are points of data. To find the direction where there is most variance, find the straight line where the data is most spread out when projected onto it. A vertical straight line with the points projected on to it will look like this:

The data isn’t very spread out here, therefore it doesn’t have a large variance. It is probably not the principal component.

A horizontal line are with lines projected on will look like this:

On this line the data is way more spread out, it has a large variance. In fact there isn’t a straight line you can draw that has a larger variance than a horizontal one. A horizontal line is therefore the principal component in this example.

Luckily we can use maths to find the principal component rather than drawing lines and unevenly shaped triangles. This is where eigenvectors and eigenvalues come in.

## Eigenvectors and Eigenvalues

When we get a set of data points, like the triangles above, we can deconstruct the set into eigenvectors and eigenvalues. Eigenvectors and values exist in pairs: every eigenvector has a corresponding eigenvalue. An eigenvector is a direction, in the example above the eigenvector was the direction of the line (vertical, horizontal, 45 degrees etc.) . An eigenvalue is a number, telling you how much variance there is in the data in that direction, in the example above the eigenvalue is a number telling us how spread out the data is on the line. The eigenvector with the highest eigenvalue is therefore the principal component.

Okay, so even though in the last example I could point my line in any direction, it turns out there are not many eigenvectors/values in a data set. In fact the amount of eigenvectors/values that exist equals the number of dimensions the data set has. Say i’m measuring age and hours on the internet. there are 2 variables, it’s a 2 dimensional data set, therefore there are 2 eigenvectors/values. If i’m measuring age, hours on internet and hours on mobile phone there’s 3 variables, 3-D data set, so 3 eigenvectors/values. The reason for this is that eigenvectors put the data into a new set of dimensions, and these new dimensions have to be equal to the original amount of dimensions. This sounds complicated, but again an example should make it clear.

Here’s a graph with the oval:

At the moment the oval is on an x-y axis. x could be age and y hours on the internet. These are the two dimensions that my data set is currently being measured in. Now remember that the principal component of the oval was a line splitting it longways:

It turns out the other eigenvector (remember there are only two of them as it’s a 2-D problem) is perpendicular to the principal component. As we said, the eigenvectors have to be able to span the whole x-y area, in order to do this (most effectively), the two directions need to be orthogonal (i.e. 90 degrees) to one another. This why the x and y axis are orthogonal to each other in the first place. It would be really awkward if the y axis was at 45 degrees to the x axis. So the second eigenvector would look like this:

The eigenvectors have given us a much more useful axis to frame the data in. We can now re-frame the data in these new dimensions. It would look like this::

Note that nothing has been done to the data itself. We’re just looking at it from a different angle. So getting the eigenvectors gets you from one set of axes to another. These axes are much more intuitive to the shape of the data now. These directions are where there is most variation, and that is where there is more information (think about this the reverse way round. If there was no variation in the data [e.g. everything was equal to 1] there would be no information, it’s a very boring statistic – in this scenario the eigenvalue for that dimension would equal zero, because there is no variation).

But what do these eigenvectors represent in real life? The old axes were well defined (age and hours on internet, or any 2 things that you’ve explicitly measured), whereas the new ones are not. This is where you need to think. There is often a good reason why these axes represent the data better, but maths won’t tell you why, that’s for you to work out.

How does PCA and eigenvectors help in the actual analysis of data? Well there’s quite a few uses, but a main one is dimension reduction.

## Dimension Reduction

PCA can be used to reduce the dimensions of a data set. Dimension reduction is analogous to being philosophically reductionist: It reduces the data down into it’s basic components, stripping away any unnecessary parts.

Let’s say you are measuring three things: age, hours on internet and hours on mobile. There are 3 variables so it is a 3D data set. 3 dimensions is an x,y and z graph, It measure width, depth and height (like the dimensions in the real world). Now imagine that the data forms into an oval like the ones above, but that this oval is on a plane. i.e. all the data points lie on a piece of paper within this 3D graph (having width and depth, but no height). Like this:

When we find the 3 eigenvectors/values of the data set (remember 3D probem = 3 eigenvectors), 2 of the eigenvectors will have large eigenvalues, and one of the eigenvectors will have an eigenvalue of zero. The first two eigenvectors will show the width and depth of the data, but because there is no height on the data (it is on a piece of paper) the third eigenvalue will be zero. On the picture below ev1 is the first eignevector (the one with the biggest eigenvalue, the principal component), ev2 is the second eigenvector (which has a non-zero eigenvalue) and ev3 is the third eigenvector, which has an eigenvalue of zero.

We can now rearrange our axes to be along the eigenvectors, rather than age, hours on internet and hours on mobile. However we know that the ev3, the third eigenvector, is pretty useless. Therefore instead of representing the data in 3 dimensions, we can get rid of the useless direction and only represent it in 2 dimensions, like before:

This is dimension reduction. We have reduced the problem from a 3D to a 2D problem, getting rid of a dimension. Reducing dimensions helps to simplify the data and makes it easier to visualise.

Note that we can reduce dimensions even if there isn’t a zero eigenvalue. Imagine we did the example again, except instead of the oval being on a 2D plane, it had a tiny amount of height to it. There would still be 3 eigenvectors, however this time all the eigenvalues would not be zero. The values would be something like 10, 8 and 0.1. The eigenvectors corresponding to 10 and 8 are the dimensions where there is alot of information, the eigenvector corresponding to 0.1 will not have much information at all, so we can therefore discard the third eigenvector again in order to make the data set more simple.

## Example: the OxIS 2013 report

The OxIS 2013 report asked around 2000 people a set of questions about their internet use. It then identified 4 principal components in the data. This is an example of dimension reduction. Let’s say they asked each person 50 questions. There are therefore 50 variables, making it a 50-dimension data set. There will then be 50 eigenvectors/values that will come out of that data set. Let’s say the eigenvalues of that data set were (in descending order): 50, 29, 17, 10, 2, 1, 1, 0.4, 0.2….. There are lots of eigenvalues, but there are only 4 which have big values – indicating along those four directions there is alot of information. These are then identified as the four principal components of the data set (which in the report were labelled as enjoyable escape, instrumental efficiency, social facilitator and problem generator), the data set can then be reduced from 50 dimensions to only 4 by ignoring all the eigenvectors that have insignificant eigenvalues. 4 dimensions is much easier to work with than 50! So dimension reduction using PCA helped simplify this data set by finding the dominant dimensions within it.

Finally I understand the concept and the use for this technique. Thank you!

Awesome Explanation!!

Thank u ! This article really helps !

This is really helpful! Thank you for explaining PCA in this way!

It does have a lot of applications in various fields.

Great explanation

I didnt understand why fig.1 has small variance and fig.2 has large one. Is it the combined distances from points to the line or the distance between the ‘arrow heads’?

It is the distance between the arrow heads. I got confused too.

I am still struggling to understand how you know if the variance is big or small from these diagrams, what do you mean by the distance between arrow heads? Any help would be much appreciated.

The arrow heads drop on the line (horizontal/vertical). Consider those as points on the line and find their variance. By looking at the vertical and the horizontal lines it is clear that the variance for the points on the horizontal line is more than those on the vertical line.

Is this basically because the horizontal line is longer than the vertical line, therefore the data points are going to be more spread out (vertically or horizontally)? Thanks for your help.

It is actually the reverse. Since the data points are more spread out on that axis, hence the length of the horizontal line is more and not the other way round.

Thank you perfectatdat, it finally makes sense! 🙂

actually the length covered on the line is depicting the variance

Variance is the spread of all the points, i.e they must be spread out/scattered. Find the mean position of all the points and then find the distance of each individual point from the mean. Square these distances to get only the magnitude of distance and not (+ or -) direction. Take the average of all the squared distances and then take the root of the final value since you initially squared them. This is your variance.

I’m just beginning to learn about eigenvectors, and this was a helpful explanation of WHY they are useful. Strict computational instruction rarely says anything about their actual usage. Thanks.

I’ve never been able to go beyond 1st paragraph of eigenvectos and eigenvalues in textbooks. This article explains concept concisely. Thanks for the post!!!!!

This is just perfect. Wish I’d found it before. Thank you!

Thanks. Awesome explanation.

Nice clear explanation!

The explanation was perfect, but I’m a bit confused with the last example… why do you say that the four pricipal components were labeled with some names in the initial dataset? Isn’t it suposed that the principal components are in another dimension, so that they are not anymore “labelled” but they are somehow a combination of those first labels??

I had the same question. My understanding is that the PCA still have same dimensions as the original data (that is 50). The PCA will be a vector with 50 weights, one for each of the dimension.

Very helpful! I wish my textbook had included this detailed explanation; it would have made it easier to understand how the dimension reduction actually occurred.

This was very helpful. Thanks alot

Best simple explanation found online! Thank you!

Extremely helpful for those of us who find it very difficult to think in terms of mathematical notation.

| There’s quite a bit of stuff to process in this post, but i’ve got rid of as much maths as possible and put in lots of pictures.

This sentence is the best of best in the post.

“They are the directions where there is the most variance, the directions where the data is most spread out.”

But why we need to find out the most variance? or say, what the most variance is useful for?

Is PCA some kind of clustering?

Thank you for this explanation. I love “for dummies” explanations, i call them “for idiots” (no offense – i am considering myself too 😉 ). Today there is much to learn, such intuitive tutorials are the only way to do it.

Great explanation!

This was great thanks, one thing I would like to know is what we can say about those 50 variables being explained in four dimensions? Using pca we lose the interpretability correct? If we are approached with a new observation of 50 levels for that same problem, we cannot simply say here are four dimensions that fit our data because these four dimensions are the eigenvectors

Simple & beautiful explanation. Thank you!

Thanks. I was searching help to teach PCA to people with no technical background and this is the best explanation I found.

It’s very useful, thank you!

best explanation of PCA, Eigenvectors, and Eigenvalues.

Great explanation! You rock.

For those wondering why the horizontal axis is the one which captures the maximum spread of the original data (“direction where there is the most variance”): http://stats.stackexchange.com/questions/2691/making-sense-of-principal-component-analysis-eigenvectors-eigenvalues

Considering that dimensions are just features ,I sometimes hear that these eigen vectors should be orthogonal.How can they be orthogonal ? I can visulaize it for 3 dimensions orthogonality ? For more dimensions what does orthogonality mean ?

Thank you so much for this explanation, it helped me a lot!

Great work!! Lucid explanation

Thank you very much! Great explanation! The best I have ever read! Would like to ask whether there is any mathematical methods (or other methods) to find the the menaings of those principal components in real life?

Very nice explanation

Reblogged this on Rami Albatal and commented:

An intuitive explanation of PCA and Eigenvectors.

This post can be an example of what I call “The art of pedagogy and communication skills”, it is very difficult to write such an intuitive and high quality explanations.

Thanks George Dallas

Really thanks for clearing up that stuff for those who do not think in “math”, you cleared up the mystery and made very simple to understand this analysis now.

I am reading about PCA for about 5 hours now ….This is the best explanation of PCA I have seen … _/\_

Thank you very much for your efforts

This was awesome! Can you please describe singular value decomposition as well.

Helpful. Thank you!! I’m just starting to learn about PCA but my classes are just throwing it out me without the conceptual explanation! This is a great start.

Fantastic Explanation! This helped me enormously.

Great explanation. This article helped me understand PCA, Eigenvectors and Eigenvalues. Thanks a lot for this superb article.

so well written!

Amazing explanations, the best I have found so far! Thumbs up!

Thank you very much for the article. I finally understood what is PCA about.