Why some portion of statistics is not used in data science

Question

I am a statistical person, I have learned all about statistics such as mean median mode, standardization, and different test such as the Z test, F test, chi-square. But when I go through the kaggle and other platforms. I hardly find data science projects using the Z test, F test, and chi-square.

I have come across most of the projects using a graphical representation of the plot to calculate the mean median, mode, etc.

My question is are these testing an integral part of data science? If so then, for what sort of problem do we use them?

Answer 1

As per my knowledge, you are looking for statistical hypothesis testing in data science, where you wanted to find out, whether there is reliability or not on two or more samples.

In kaggle, you won't find the hypothesis testing problem in huge numbers. This is the reason why you don't find people who use the Z test, F test, and chi-square. You can only find the supervised and unsupervised learning problems, where you use different kinds of the algorithm to predict the target variable and find the accuracy of the error rate.

I prefer to learn both statistical hypothesis testing and also the machine learning techniques

You can use the link Data Science for more information on the subject and improve your knowledge