Today we will see what is inferential statistics and why do we need it? So, if we do any data analysis then we require large number of data or large amount of data.
But to collect the large amount of data we might also require huge time.
We can even require a lot of money and with the help of lot of resources, we are able to collect a large amount of data, which is not feasible.
So, what we do is we take a smaller sample of the data and through that we do our analysis on that particular sample.
With that sample for the entire population data, we do the analysis and find the result so that our efforts are saved.
So, let's understand this with an example.
Suppose that Amazon they have one sales team, that sales teams have to find out that the products that they are selling, how many of those products are defective products? But you think about it, Amazon delivers so many products, if we start to analyse the data of all those products, then that will be time consuming for us.
Plus, we will require more labour.
So, what we do is, instead of using the complete population, we take a small sample from it.
Suppose we have taken 1000 products.
Okay, created a small sample and over that sample we have done the data analysis and we have found out that how much promotion of defective product are there in it or er can even call it as defect rate.
We found out that defect rate of the sample, we inferred the defect rate of all my products.
What happened with it? We saved a lot of energy plus we got to know the full population’s data and the full population’s defect rate through our sample.
So, the process of what we infer inside through the sample data, we call that as inferential statistics.
Where has this term inferential come from? Since we infer the sample data and find the population data.
Inferential statistics is a very important field.
It is used in manufacturing industries like pharmaceuticals, food industry, and almost all large industries.
Why? Because we have bulk data present in it and we cannot do analysis on all the data.
For that we want to take a small sample of data and then draw the inferences about the population.
This is also used a lot in our technology companies like Google and Amazon.
Suppose that particular company gets a new feature, to test those features, what they do is they test different hypotheses to show how effective their particular product is, they can give us the information about it.
For example you must have seen that what’s app or any other like Meta Facebook in today's day and age.
When they get any new feature, they give a beta version.
What is basically that beta version that we will be seeing ahead.
By doing AB tests all these big companies do their analysis and they get to know about the effectiveness of their product features.
Now we have seen why inferential statistics are useful for us.
After using inferential statistics we will only get an estimate of the population data.
Are you understanding? Because the data that we have is not complete.
So, we will not get the exact values.
We will not get the exact values.
What does it mean by we will not get the exact values? We will have to increase certain estimates for a limited level of certainty.
This means that with the data that we have, we have to work on levels of certainty.
For that we have to learn probability and concepts related to that.
If you have any queries or comments, click the discussion button below the video and post there. This way, you will be able to connect to fellow learners and discuss the course. Also, Our Team will try to solve your queries.
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