Box-Jenkins Analysis

This part covers the full Box-Jenkins workflow for equi-distant quantitative time series: identification, estimation, diagnostics, and forecasting. The objective is to build models that are statistically adequate and operationally useful.

Compared with the previous time-series introduction, this section is more formal and model-centric. You will move from visual heuristics to explicit ARIMA specification, residual testing, and scenario-oriented forecasting.

How This Part Is Structured

Block	Purpose	Main chapters
Conceptual setup	Introduce Box-Jenkins principles and theoretical building blocks	Introduction to Box-Jenkins Analysis, Theoretical Box-Jenkins Analysis
Stationarity and model identification	Diagnose and induce stationarity, then choose candidate orders	Stationarity and Models, Identification of ARIMA Models
Estimation and forecasting	Fit ARIMA candidates and generate forecasts	Estimation of ARIMA Models, Forecasting ARIMA Models
Dynamic interventions and cross-series effects	Model external shocks and inter-series influence	Intervention Analysis, Cross-Correlation Function, Transfer Function Noise, General-to-Specific

How To Use It

Use this part when your objective is explainable short-term forecasting or dynamic impact analysis.
For rapid method routing, combine this with Appendix A — Method Selection Guide (Appendix A), then document final model choices and diagnostics in the same way across chapters.