Alright, so again, let me, I'm sorry for studying out too late.

So we're at the second exercise.

So exercises or tutorial if you want for lecture number nine, that number eight.

It's still number eight because we didn't finish the tutorial last time.

So number eight is a last last time we talked about testing of hypotheses and we were just

looking at a few basics.

So that's where we are preliminaries.

Why is the normal distribution so interesting?

So back to the call lab, I had showed you a few examples of code.

Let me make this larger.

It's showed you a few examples of code which would generate interesting graphs and especially

we talked about the uniform distribution which is something coming out of, say, throwing

a dice and a high number of times and then wondering how often does each face appear.

And as we would expect each face appears, well, the total number of of roles divided

by the number of faces, so there's going to be 60,000 divided by six which makes it for

10,000.

And as you have seen, so I hope some of you at least have worked out the examples, the tasks

that I said here.

If you increase the number of times that you roll your dice, then the frequency's appear

to be really similar to one another.

So that's a uniform distribution.

So main point being there's no pre-deleted outcome.

So all the faces are supposed to come out with the same likelihood unless the dice is loaded

which is not.

And so you get this thing which is the uniform distribution.

Now in many other cases, oh and by the way, there's this little thing about the bins.

You also find interesting.

You see the ones giving you messages in the chat.

The video is stopped and the first slide is being showed.

Sasan, do you see what I'm showing here?

I'm guessing it was a glitch on your hand on your site because you went out of the meeting

and then came back in.

Yes, I'm so sorry, I think it's my side.

Okay, right.

All the others are seeing me scrolling the screen.

I'm assuming that.

All right.

So what was that about?

Yeah, exactly.

So uniform distribution actually numbers from 0 to 1.

So this is a generalization to the continuous case.

And remember that in order to plot the real outcome of something, you need to decide a number

of bins.

We have 100 bins in which we divide the interval from 0 to 1 and then we have 10,000 numbers

drawn from a uniform distribution and that means that each bin should have at least more

or less 100 out of them.

So the way you write down the uniform distribution is like this, uniform of A, B, which means

that it's basically some constant value between 0 and 1 in this case because A is 0 and B

is 1.

And outside this interval, it will be 0.