Time series analysis What is the basic principle of time series analysis?

ARIMA model (moving average autoregressive model), which is the most commonly used method for time series prediction and analysis. Historical data can be used to predict the future situation. ARIMA model can be divided into three items, namely AR model, I-difference model and MA model. SPSSAU intelligently finds out the best AR model, where I is differential value and MA model, and finally gives the best model prediction result. The principle of finding the best model intelligently by SPSSAU is to use the minimum AIC rule to traverse all possible model combinations to build the model, and finally get the best model by combining the minimum AIC rule.

Of course, researchers can also set their own AR model, differential order and MA model, that is, set autoregressive order P, differential order D and moving average order Q respectively, and then build the model. As for the autoregressive order P, the difference order D and the moving average order Q should be set appropriately. It is suggested that researchers use the partial (auto) correlation diagram for analysis (SPSSAU also intelligently provides suggestions on P or Q) and use ADF test analysis to get the appropriate difference order D (SPSSAU also intelligently provides suggestions on the best difference order D).

ARIMA model can be divided into three items, namely AR model, I-difference model and MA model. SPSSAU intelligently finds out the best AR model I, namely differential value and MA model. Of course, if the researcher sets the AR model, differential order and MA model, that is, sets the autoregressive order P, differential order D and moving average order Q respectively, then SPSSAU will build the model according to the researcher's setting. It is suggested that users directly use the intelligent analysis of SPSSAU.

Spssau operation is as follows: