3 mo treasury yield borrowing costs Dow industrials NY Times 18 Sept 2008 front page
Stat Sept 2008 D. R. Brillinger Chapter 3 mean function variance function autocovariance
Strictly stationary Normal/gaussian - all joint distributions jointly normal Wide sense stationary vs. second-order
Properties of autocovariance
Useful models Purely random Building block
Random walk not stationary
*
Moving average, MA(q) From * stationary
Backward shift operator Linear process. Need convergence condition
autoregressive process, AR(p) first-order, AR(1) Markov Linear process For convergence/stationarity *
a.c.f. From * p.a.c.f.
In general case, Very useful for prediction
ARMA(p,q)
ARIMA(p,d,q).
Yule-Walker equations for AR(p). Correlate, with X t-k, each side of For AR(1)