16 commonly used data analysis methods-time series analysis

A time series is a numerical sequence in which the observed values ??of a certain variable in the system are arranged in chronological order (with the same time intervals) to display the change process of the research object within a certain period of time and to find and analyze the changing characteristics and development of things.

Trends and patterns.

It is the total result of a certain variable in the system being affected by various other factors.

The main purpose of studying time series is to make predictions and predict future changes based on existing time series data.

The key to time series forecasting is to determine the pattern of change in an existing time series and assume that this pattern will continue into the future.

The basic characteristics of time series: It is assumed that the development trend of things will extend into the future. The data on which the prediction is based is irregular. The causal relationship between the development of things is not considered. Time series data is used to describe the characteristics of the development and change of phenomena over time.

Time series considerations Time series analysis can be divided into traditional time series analysis and modern time series analysis based on its development history stage and the statistical analysis methods used. Depending on the observation time, the time in the time series can be years,

Quarter, month or any other time format.

The main factors to consider when analyzing time series are: l Long-term trend? Time series may be quite stable or show a certain trend over time.

Time series trends are generally linear, quadratic or exponential function.

lSeasonal variation (Seasonal variation) changes over time and presents a sequence of repetitive behaviors.

Seasonal changes are often related to date or climate.

Seasonal changes are often related to annual cycles.

l Cyclical variation Relative to seasonal variation, time series may experience "cyclical variation".

Cyclical changes are usually due to changes in the economy.

l Random effects In addition, there are also accidental factors that affect the time series, causing the time series to show some random fluctuations.

Accidental fluctuations in a time series after removing trends, periodicity and seasonality are called randomness, also known as irregular variations.

The main components of time series The components of time series can be divided into four types: l Trend (T), l Seasonality or seasonal changes (S), l Periodic or cyclic fluctuations (C), l Randomness or irregular fluctuations (I

).

One of the main contents of traditional time series analysis is to separate these components from the time series, express the relationship between them with certain mathematical relations, and then analyze them separately.

Basic steps for time series modeling: 1) Use observation, survey, statistics, sampling and other methods to obtain time series dynamic data of the observed system.

2) Make a correlation diagram based on dynamic data, conduct correlation analysis, and find the autocorrelation function.

Correlation graphs can show changing trends and cycles, and can find jump points and turning points.