T10 - − + = − − 2 1 5 . 1 . 1 5 11 , 9 2 1 = = Z Z...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon
STA4005B Time Series (2007-2008) Tutorial 10 (1/4, 2/4) (Time & Venue: T9, BMS LT; W5, MMW 704) Forecasting Def: Given a time series the MMSE (Minimum Mean Squares Error) forecast at time t with l is , , , , 2 1 t Z Z Z K ) , , , | ( ) ( ˆ 1 1 Z Z Z Z E l Z t t l t t K + = and the l -step ahead forecast error is given by . ) ( ˆ ) ( l Z Z l e t l t t = + Example 1 Random Walk with Drift Given and 1 1 a Z = t t t a Z Z + + = 1 0 θ K , 4 , 3 , 2 , = t Find , ) ( ˆ 1 l Z 1 l 1
Background image of page 1

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

View Full DocumentRight Arrow Icon
Forecasting with General Linear Process For any ARMA model, we can be rewritten the series in the following form: L L + + + + + + = + + + + t l l t l t l t l t a a a a Z ψ µ 2 2 1 1 as 0 ) , , , | ( 1 1 = + Z Z Z a E t t i t K for all , so 0 > i L + + + + = + + 2 2 1 1 ) ( ˆ t l t l t l t a a a l Z and 1 1 2 2 1 1 ) ( + + + + + + + + = t l l t l t l t t a a a a l e L Example 2 Given an AR(1) model: t t t a Z Z + = 1 φ , find the value of . ) ( l e t 2
Background image of page 2
Updating the ARIMA Forecasts If new information arrives, we need to re-calculate the forecasting L + + + + = + + + + 2 3 1 2 1 ) 1 ( ˆ t l t l t l t a a a l Z ψ µ L + + + + = + + 2 2 1 1 ) ( ˆ t l t l t l t a a a l Z Then L + + + + = + + + + 1 2 1 1 1 ) ( ˆ t l t l t l t a a a l Z ) ( 1 2 1 1 L + + + + = + + + t l t l t l a a a Thus ) 1 ( ˆ )] ( ˆ [ ) ( ˆ 1 1 + + = + + l Z l Z Z l Z t t t l t Example 3 Suppose and t t t t a Z Z Z +
Background image of page 3

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

View Full DocumentRight Arrow Icon
Background image of page 4
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: − + = − − 2 1 5 . 1 . 1 5 11 , 9 2 1 = = Z Z and 10 3 = Z . Find and . ) 1 ( ˆ 3 Z ) 2 ( ˆ 3 Z If what is ? , 12 4 = Z ) 1 ( ˆ 4 Z 3 Prediction Interval Since )]) ( [ , ( ~ ) ( ) ( ˆ 1 l e Var N l e l Z Z t t t t = − + The )% 1 ( 100 α − prediction interval for is given by l t Z + ) )) ( ( ) ( ˆ , )) ( ( ) ( ˆ ( 2 2 l e Var z l Z l e Var z l Z t t t t + − Example 4 Given and . Find the 95% prediction interval for 2-step ahead forecast at t =1 for the ARMA(1, 1) model 5 . , 2 1 1 = = a Z 1 2 = a σ t t a B Z B ) 3 . 1 ( ) 1 )( 2 . 1 ( − = − − 4...
View Full Document

This note was uploaded on 11/02/2011 for the course STAT 4005 taught by Professor Wu,kaho during the Spring '08 term at CUHK.

Page1 / 4

T10 - − + = − − 2 1 5 . 1 . 1 5 11 , 9 2 1 = = Z Z...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online