This page is the note of the book: introduction of TS and forecasting (from Brockwell), now I have stoped reading this book. To resume the progress, please check ‣
After subtraction or difference, if the sequence becomes iid noise, then the model accomplish; otherwise, we should build models by the theory of stationary process (Chapter 2).
(a) The sample autocorrelation function 【ACF】:
sample autocorrelations up to lag 40 andfind that more than two or three values fall outside the bounds $\pm \frac{1.96}{\sqrt{n}}$
(b) The portmanteau test
(c) The turning point test
(d) The difference-sign test
(e) The rank test
(f) Fitting an autoregressive model
(g) Checking for normality by Gaussian-QQ plot
2.1 Basic Properties P49.eg
2.2 Linear Process
2.3 Introduction to ARMA Process
2.4 Properties of the Sample Mean and Autocorrelation Function (Need reread)


variance of Xbar_n



2.5 Forecasting Stationary Time Series
2.6 The Wold Decomposition