Then you actually estimate the quantity of interest that is,

as I gave the example, what is the effect of a specific ad on your sales?

The next step in hypothesis testing is to calculate your test statistic.

That is the observed value of the statistic which you want to test.

This can take different distributions, like a T distribution, a F distribution,

a distribution and so on.

Finally, once you have this test statistic calculated,

you evaluate the P-value of the hypothesis test.

Now, what is a P-value?

P-value essentially measures the probability that an observed effect

is there in the sample given that your null hypothesis is true.

So again, going back to our previous example, a P-Value in our example can be,

what is the probability that there is an observed effect of the ad in our sample.

Given that,

the ad has no effect because remember a null hypothesis is the ad had no effect.

Now, the next step in our hypothesis testing process

is to arrive at the decision.

So how do we arrive at the decision?

Basically, we reject the null hypothesis and

accept the alternative hypothesis If our p value takes the value greater than zero.

Now when you do a hypothesis testing, there can be two types of errors.

The first type of error is called Type I error, which is obvious right?

So basically what is Type I error?

Type I error is the probability that you reject the null hypothesis given that

your null hypothesis is actually true.

Again, back to our example, so, a type I error can be what is

the probability that you select an ad when the ad has actually no effect.

So, that's an example of type I error.

The second type of error is called type two error.

Here, we actually do not reject the null hypothesis

even though the null hypothesis is false.

So again, going back to our example, this is where you actually do not

select an ad even though the ad has some effect on your sales.

So that is mainly the overall preview of hypothesis testing and in the next

video we are going to dig deep into some more about data analysis fundamentals.

So talk to you soon.

Thank you.

[MUSIC]