Statistics is a comprehensive science that searches, organizes, analyzes, and describes data in order to infer the nature of the object being measured, or even to predict the future of the object.

The central problem of statistics is how to explore the real situation about the whole according to the sample. In the past, when we had limited data and limited arithmetic, the role of statistics was important.

With information technology, the amount of data we have access to is getting bigger and bigger, our computing power is getting stronger and stronger, and the historical mission of statistics seems to be reaching its end.

In the era of big data, we have access to full samples, and in the era of big data, we have different ways to work with the data and to understand all aspects of the sample. Compared to statistics, the sample specification, the total amount of data is limited, and the observation of a single target, in this case, statistics is applicable. However, the case of too much data, for example, is the network massive data, how do you sample, how do you observe it?

One of the characteristics of big data is diversity, there is a certain degree of correlation between different sources, different dimensions of the data, can be cross-validated, the use of big data to make decisions, the decision will be able to move from a crude to intensive.

So, statistics seems to have no value?

Big data is the full sample, yet some data is not of great value and can even be misdirected.

Google's founders, Larry. Page and Sergey . Brin, when interviewing for a job, ask for a candidate's SAT scores from college and their grade point average when they graduated from college.

However, every time those 40-something, accomplished professional managers were asked about their grades, they were confused.

They base this on the fact that the SAT shows talent and the college grade point average shows achievement.

Does performance have anything to do with those grades? It doesn't.

This data, while all-inclusive, is still subject to uncertainty.

?

Big data informs but does not interpret information, which is the nature of big data correlations.

Just like the stock market, even if all the data is published, people who do not understand still do not know the information represented by the data. In the era of big data, statistics is still the soul of data analysis. As Prof. Michael Jordan of the University of California, Berkeley points out, "Big data research without systematic data science as a guide is like building bridges without utilizing knowledge of engineering science; many bridges may collapse with serious consequences."

In the era of big data, we can't ignore some of the new methods and ideas, but many of the fundamental issues of data analytics are not fundamentally different from the era of small data.