Original subtitles

In this lesson.

I wanted to take a little bit of time to speak about correlation analysis.

So correlation analysis is when you've got two variables and you're wanting to see if there is a correlation

between the two variables.

So if one variable goes up, does the other variable also go up?

Sometimes you can have a negative correlation as well where you can have an item going up, but the

other variable then goes down.

So we're going to have a look at our scatter chart and just see some of the functionality that we've

got available to us when it comes to analyzing your data.

Again, you might find this very useful when you when you're working with your own data and you want

to take to variables and see if there is actually correlation between the two.

Now, please note in this training data it is actually quite highly correlated.

So you're going to see that there is a very strong correlation.

But I just wanted to really show you how the scatterplot can work and also some of the benefits you

can get from this type of analysis.

So let's move on and we're going to, first of all, go across the scatter chart and we're going to

pick this visualization like we've done previously.

What we're going to do is we're just going to select it, make a nice and big, and you see that you

get your values X and Y axis as options that you might want to work with.

Now, in this case, what we're going to do is we're going to take our X axes and we're going to say

we want to see our sales values on there.

Now, what you can see is it's taken one value, which is my total sum of sale.

Then in the Y axis, we want to say we want to look at our profit.

And now what it's done is it's now taken one data point and it's taken the total sales and the total

profit, and it's created that data point for everything that is in the data set.

So really what we want to do from here is we want to break this out a bit.

We want to be able to see this from different aspect.

So what we're going to look at is our product pedigree.

So we're going to take our product category, we're going to drop it into our values.

So now you can see that it's actually created a data point for each of my product categories that we

have.

And you can see now it's taken the sum of the sales and the sum of the profit, and it's basically now

plotted those into this graph.

And what you can see from this is you actually do have a sort of a 45 degree line going upwards, which

shows that the strong correlation because what it means is that as your sales is going up, so is your

profit.

So that's basically what we're seeing with this correlation.

So over here we've got quite high sales and we can also see that there is quite high profit.

And that's really what we're looking at from the scatter chart is to understand is there a correlation

from these two variables?

And quite clearly there is.

We could look at this in more detail.

For example, you could take more data points so we could say, let's have a look at our product subcategory.

So you can see now we've got a product subcategory and you can see again that there's very strong correlation

between these two.

Now, if we had values that were set over here, then this is probably items that you might want to

look at in more detail because this is items where you got a large amount of sales, but you've got

very little profit.

However, if you have items that are over here, this would be good for you because the sales value

is small, but the profit value is quite high.

So you obviously making quite high profit from low sales values there.

So again, you can see that we've got these dots now, each one representing a different product subcategory.

Now we know that we've got quite a lot of product names and you could actually pop this in there as

well.

Because remember, what we're trying to find here is we're just looking for anomalies or we're looking

for things that don't fit into the trend.

And again, most of these are actually fitting into the trend.

Again, we can see that there's a 45 degree line basically showing the options here.

That's the first part that we can look at from here.

There are some formatting options that you could ever look at here.

Very much kind of like what we've seen previously, and you've got your x axis, your y axis options

that are pretty much exactly the same.

You've got grid lines.

We've also got markers.

If you want to put a marker onto onto your options, it probably wouldn't make too much sense here.

This category label.

I want to leave until a little bit later.

We're going to look at something called the bubble plot.

And I'm going to show you how you can change the settings just to allow us to see correct categories

on this.

In terms of general as well, you see that there's not too much different new properties title effects,

the ones that we're used to.

However, if we go across to our analytics, you will see that there is some analytics that we can see

from here.

So we can pick a trend line and this can be very useful to see what that actual trend is.

And as I'm saying, we've got a 45 degree line which is showing me that as the sales get higher, so

does the profit.

Now this can be quite useful.

If you wanted to pair this, for example, with say that you wanted to have a slicer now because this

is my product names, what I could do is I could actually use my product category and I could see how

all my different product names and my product categories performing.

So if I drop that in there, I could see if there's any noticeable difference between the trends between

these different product categories.

As you can see in this data, pretty much there is an all pretty good sales goes up, so does profit

go up as well?

But that's just one of the options that you could use, is to pair it with a slicer and to be able to

use your trend line.

You will see that there's other options as well.

You could turn on your symmetry shading and would allow you to play with that.

Can see looking at your upper shading, lower shading options ratio lines as well.

You want to have a ratio line in there that shows you the ratio that is on there.

So as I say, these these are options for you to be able to play with.

Now I want to do want to move to is just another option, which is called a bubble plot.

So I'm going to remove the trend line just to make this simpler.

And what we're going to do is we're going to go back to our product category because a bubble plot,

because we're going to use the size of the bubbles to represent something.

So basically, in this case, we're going to use order quantity.

So let's pop our order quantity into our size.

And now you're going to see that the size will now change according to how much order quantity.

So basically we're using a third variable.

But what we can do is we can go to our formatting and you can also turn your category label on here.

So basically now you can see the name of the product category and you can basically now see the different

product categories and with the labels on and also with the bubble size, that will show you your order

quantity.

So this is called the bubble plot.

What you do get is also an option that you can actually show a visualization of this changing of a ton.

Okay.

So one of the options that we've got is we can take our order date and we can actually drop it into

our player axis.

So when we drop it in here and press play, it will actually show you day by day.

Now what is actually happening with our products?

Now, another option you could try with this is to try drop down.

Go to our data hierarchy and you'll see that it's currently can be changed to year.

So this is just a different view.

And with this now we press and play and then you can sort of see year by year what the changes have

been.

So as I said, this is just a different view.

It just allows you to do a bit of correlation analysis with your data.

You can look at different values in your X and Y axis and see how they're correlated.

And then there are the tools such as the player axis and the size where you can actually ever look at

your data from a different perspective.

Okay.

We're going to conclude the lesson there.

We'll see you in the next one.