This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Joint distribution = f(x1, x2, …, xt) everything about the random variables Check for weak stationary: mean doesn't change over time variance does not change covariance doesn’t change over time (does not depend on time) implies that the variance is going to be stable (the same over time) As long as the lag is the same, the difference in variance should be the same second order stationary (covariance stationary) only concentrate on mean, variance, and covariance to see if it is weakly stationary gamma = covariance gamma(t + h, t) = E(Xt+h - mu) * (Xt - mu) = E(Xh - mu) * (X0 - mu) gamma (h, 0) autocorrelation = gamma(h) / gamma(0) cov(1,2) = cov(2,1) gamma(0) = cov(xt, xt) = var(xt) = sigma^2...
View Full Document
- Fall '13
- Variance, Probability theory, Xt, WT, GAMMA, Xs Xt