{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

3 - Introduction to Time Series Analysis Lecture 3 Peter...

Info icon This preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
Introduction to Time Series Analysis. Lecture 3. Peter Bartlett 1. Review: Autocovariance, linear processes 2. Sample autocorrelation function 3. ACF and prediction 4. Properties of the ACF 1
Image of page 1

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Mean, Autocovariance, Stationarity A time series { X t } has mean function μ t = E [ X t ] and autocovariance function γ X ( t + h,t ) = Cov ( X t + h ,X t ) = E [( X t + h - μ t + h )( X t - μ t )] . It is stationary if both are independent of t . Then we write γ X ( h ) = γ X ( h, 0) . The autocorrelation function (ACF) is ρ X ( h ) = γ X ( h ) γ X (0) = Corr ( X t + h ,X t ) . 2
Image of page 2
Linear Processes An important class of stationary time series: X t = μ + summationdisplay j = −∞ ψ j W t j where { W t } ∼ WN (0 2 w ) and μ,ψ j are parameters satisfying summationdisplay j = −∞ | ψ j | < . 3
Image of page 3

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Linear Processes X t = μ + summationdisplay j = −∞ ψ j W t j Examples: White noise: ψ 0 = 1 . MA(1): ψ 0 = 1 , ψ 1 = θ . AR(1): ψ 0 = 1 , ψ 1 = φ , ψ 2 = φ 2 , ... 4
Image of page 4
Estimating the ACF: Sample ACF Recall: Suppose that { X t } is a stationary time series. Its mean is μ = E [ X t ] . Its autocovariance function is γ ( h ) = Cov ( X t + h ,X t ) = E [( X t + h - μ )( X t - μ )] . Its autocorrelation function is ρ ( h ) = γ ( h ) γ (0) . 5
Image of page 5

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Estimating the ACF: Sample ACF For observations x 1 ,...,x n of a time series, the sample mean is ¯ x = 1 n n summationdisplay t =1 x t . The sample autocovariance function is ˆ γ ( h ) = 1 n n −| h | summationdisplay t =1 ( x t + | h | - ¯ x )( x t - ¯ x ) , for - n<h<n . The sample autocorrelation function is ˆ ρ ( h ) = ˆ γ ( h ) ˆ γ (0) . 6
Image of page 6
Estimating the ACF: Sample ACF Sample autocovariance function: ˆ γ ( h ) = 1 n n −| h | summationdisplay t =1 ( x t + | h | - ¯ x )( x t - ¯ x ) . the sample covariance of ( x 1 ,x h +1 ) ,..., ( x n h ,x n ) , except that we normalize by n instead of n - h , and we subtract the full sample mean.
Image of page 7

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}