Before testing any hypothesis testing, we will have to create one null and one alternate hypothesis.
So, first we will see what is null hypothesis, null hypothesis is called my status quo.
Which means, whichever is my current state, my current belief about the population that will be called my null hypothesis.
This also means that if there is no difference in my old situation and in my current situation then we will be define it with null hypothesis.
We did denote null hypothesis with H knot.
Second is my alternate hypothesis.
Alternate like it's named suggests that this is opposite of my null hypothesis.
If I have created my null hypothesis, to challenge it or to oppose it, if we create a second hypothesis, that will be called my alternate hypothesis.
So, basically the challenge to the status quo is called alternate hypothesis.
We denote it by H1.
Lets understand null and alternate hypothesis through one example.
How do we formulate it and how it is used.
So, let's imagine that there is a person on whom murder charge is levied.
The murder charge means that, that person will go to the court.
The jury can prove him innocence or it can prove him guilty as well.
Which means if he's proved innocent, that will mean that he has not committed that murder.
And the jury has accepted that he has not committed the murder.
But if he is proven guilty then this means that he has committed the murder and he will go to the jail.
So, in this situation, even now there is a charge levied on him.
But earlier there was no charge on him.
So, our status quo will be that the defendant is innocent.
So, the null hypothesis that we will formulate, it will be that defendant is innocent.
What will be the alternate hypothesis in this situation? Since in the null hypothesis we have taken the defendant as innocent.
So, in alternate hypothesis my defendant will be called guilty.
So, in this way we have created null and alternate hypothesis but will be get the outcome just by creating it? What is ther decision taken by the jury? When we once create our alternate and null hypothesis.
Now we will see how we accepted null hypothesis or we failed to reject null hypothesis.
So, now the jury has to take a decision.
Let’s assume that jury’s first decision is that my defendant is guilty, guilty means that what was my null hypothesis, that he is innocent.
So, in this situation it so happened that we rejected null hypothesis.
Why? Because the defendant was found guilty.
In this situation we will say that our result went in favour of alternate hypothesis and we rejected null hypothesis.
So, basically what must have happen? Jury must have got enough evidence that he proved his claim and proved the defendant guilty.
What can be the second scenario? That the jury freed the defendant.
This means that the jury does not have enough evidence that he can speak in support of alternate hypothesis.
My alternate hypothesis was that the defendant is guilty but since we did not have enough evidence, the jury freed him but there is one thing to pay attention is that jury has not called him innocent.
He did not have enough evidence to speak in favour of alternate hypothesis.
That’s the reason why he freed him.
So, basically we never say that we have accepted null hypothesis, in this situation if you see ideally we feel that we have spoken in favour of null hypothesis and we have accepted but we never say this, we always say that we have failed to reject the null hypothesis.
So, basically there are 2 outcomes of my any hypothesis testing.
A is either we reject the null hypothesis, second is we fail to reject null hypothesis.
So, we never say that we say that we are accepting our null hypotheses.
So, this means that we don't have enough sufficient evidence that we can talk in favour of alternate hypothesis.
That’s why we call it fail to reject null hypothesis.
So, basically, we saw in this entire example that how we create a null hypothesis and one alternate hypothesis.
On those two hypothesis how outcomes come of hypothesis testing.
Like if the defendant is guilty that will be my rejection of the null hypothesis.
When the defendant is not guilty means that is called as fail to reject null hypothesis.
In this way based on the different context, in different use cases, we apply hypothesis testing and find out that how far is my claim valid.
Now we will get many such example where looking at the statement we cannot say that this would be my null hypothesis or alternate hypothesis.
Because many of the problem statements that you will see or which we use in our day to day life, it is formulated by all the terms like, at least, at most, less than, greater than.
But what happens in this scenario that looking at the words we cannot say that what is my null hypothesis or what is my alternate hypothesis.
So, for this it is important that we create one thumb rule, which means we will create a trick with which we will get to know that what will be my null hypothesis and what will be my alternate hypothesis, looking at our problem statement.
So, wherever you are getting is equals to, greater than equal to or less than equal to sign.
So, that formulates my null hypothesis.
Wherever you will get not equal to, greater than or less than symbol, that will create my alternate hypothesis.
I will explain it with an example, suppose we take 2 situations.
In the 1st
situation, Flipkart has claimed that their total turnover on 2016 December,
which means the money that they, their total earnings were around $14 billion.
Now, this is their claim and here we are using at least term, so what would be my null hypothesis, the total valuation of Flipkart is greater than, is equals to $14 billion, because we just now saw that to create the null hypothesis, we use greater than equal to sign.
So, its opposite means alternate hypothesis would be total valuation less than $14 billion.
If I see another situation, that the Flipkart claimed their total valuation, which means that total earnings in 2016 December was greater than $14 billion.
Now we can see here very clearly that is equals to is not used.
So, this means that the greater than equation will create my alternate hypothesis, which means total valuation is greater than $14 billion, that will be my alternative hypothesis and its opposite, which means total valuation is less than equals to $14 billion, will create my null hypothesis.
So, in this way, in different scenarios we can use at least, at most, less than, greater than symbols, we can create our null and alternate hypothesis.
If you have any comments or questions related to this course then you can click on the discussion button below this video and you can post them over there.
In this way, you can connect with other learners like you and you can discuss with them.
This course is really nice, just have one question in empirical rule explanation , SD deviation example trainer is saying mean however mean (20+30+40+50+60+70/6) value is different kindly confirm than