00:01
We have a set of 10 cards.
00:03
Three of them are red, seven of them are black.
00:05
We put them down and we start turning them over one at a time until we get to the first red card.
00:11
And we want to know the distribution of x, the number of cards we have to turn over.
00:22
Okay, is x a normal distribution with mean three, a binomial distribution, the geometric distribution, uniform, or none of the above? so let's go through some of the requirements here, when you would see these.
00:40
First of the normal.
00:41
First of all, this cannot be normal.
00:43
The normal distribution is continuous.
00:46
It can take any value.
00:48
X is discrete.
00:49
It can only take whole numbers.
00:51
It's either going to be one, two, three, up to ten.
00:55
If it take, well, not ten, really.
00:57
There's three red cards.
00:58
But it has to take whole numbers.
01:01
So it can't possibly be normal.
01:03
Besides, the normal distribution looks like this.
01:07
It's symmetric, it's unimodal, you see it for things where you have lots of tiny factors influencing something, like height or weight, or sample distributions.
01:19
B, a binomial.
01:20
So a binomial has a fixed number of independent trials, and you're looking at the number of successes...