Now here comes the quiz.

But there is another measure for such correlation, and it was also invented by Pearson.

So the higher the value in 1 diagonal and the lower the value in the other diagonal, the stronger the correlation.

And we start, of course, with the strongest correlation.

phi

Phi

phi

This is then normalized according to the square root of the sums of the columns and rows.

Well you might wanna become a number theorist.

You also compute sums over all rows as well as sums over all columns as well as the total sum of events.

One important function he defined is called the 'phi function.'

Modulus

Cell phones, online dating, WiFi.

And that's true of any sample statistic.

Correlation

Extinction

Cooling factor

So I can just rewrite that as A so now A is not a function anymore, it's just a number A times the second derivative of y, with respect to x, plus B times the first derivative, plus C times y.

Here we have the product of the first diagonal minus the product of the second diagonal, and here we have the square root over the product of the individual sums of the rows and columns.

AMS Binomial Expression

Scale by factor

n Choose m

Next I want to focus for a little while on the notion of a clustering coefficient.

Wireless interlinking.

This is supposed to be a constant

So there's a fair amount of interconnection between a node's neighbors.

So just getting it correct to a couple of digits is probably sufficient.

2nd derivative r squared.

Now it's about eighteen.

Normal Binomial

Displaystyle Binomial

Textstyle Binomial

So we'd like to have a way of getting a pretty good answer in time a lot less than this.

We'll come back to that again in multiple regression, and in lecture nine in R. And then, the importance of this distinction between unstandardi zed and standardized regression coefficients.

How strong is the relationship?

But again, now that we're in multiple regression, this is not the same as the correlation coefficient.

They're on the same side of the equation.

So let's multiply both sides by negative 3, right?

Either way is a perfectly good series, and it's got another similarity with the Fibonacci series.

Let's just solve x squared minus x minus thirty is equal to zero.

I think you see what I did.

So, I just said and look, the a is just the coefficient on the x term, right? a is the coefficient on the x squared term. b is the coefficient on the x term, and c is the constant.

You literally just look at this coefficient right here, and you say, OK, well what's half of that coefficient?

When we complete the square, we just take half of this coefficient.

Low error factor