Descriptive analysis is an important step in data analysis. Before descriptive statistical analysis, we should first understand the necessity of collecting data, analyzing data and identifying some common data sources. Then, we should understand the common data types and methods of data summary in practice; Finally, the single variable numerical description method and two or more data analysis methods are determined.

1.？ Data: definition and objectives

First of all, we should make some definitions clear.

Data: used to show and explain the facts and figures collected, analyzed and refined;

Variable: a symbol or indicator that can take different values. Such as: industry, stock price, market value;

Decision variables: the value of variables is directly controlled by decision makers;

Random variable/uncertain variable: the value of the variable is not affected by the factors directly controlled by the decision maker, and there may be uncertainty fluctuations;

Observed value/observed value: a set of values corresponding to a set of variables;

Descriptive analysis, that is, through the analysis of the collected data, to gain a good understanding of the variation and its impact on the business environment.

2.？ data type

(1) population data and sample data: In many cases, it is not feasible to search for data from the population (a collection of elements of interest). At this time, data can be collected from a subset (sample) of the population. It is very important to find sample data that can represent the whole population. Only in this way can those sample data be extended to understand the overall situation.

(2) Quantitative data and attribute data: Quantitative data refers to data that can perform numerical and arithmetic operations such as addition, subtraction, multiplication and division, such as the market value of a company; Attribute data refers to data that cannot be arithmetically operated. Descriptive analysis of these data can only count or calculate the proportion of observed values in various categories, such as the industry to which the company belongs.

(3) Cross-sectional data and time series data: Cross-sectional data refers to the data of some individuals collected at the same time or almost at the same time; Time series data: refers to the data of several periods. Time series data graphs can help analysts understand what happened in the past, identify trends that change with time, and predict the future.