ebook img

Lecture 2: ARMA(p,q) models (part 3) - unice.fr PDF

32 Pages·2011·0.47 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Lecture 2: ARMA(p,q) models (part 3) - unice.fr

Lecture 2: ARMA(p,q) models (part 3) Florian Pelgrin UniversityofLausanne,E´coledesHEC Departmentofmathematics(IMEA-Nice) Sept. 2011 - Jan. 2012 FlorianPelgrin (HEC) Univariatetimeseries Sept.2011-Jan.2012 1/32 Introduction Motivation Characterize the main properties of ARMA(p,q) models. Estimation of ARMA(p,q) models FlorianPelgrin (HEC) Univariatetimeseries Sept.2011-Jan.2012 2/32 Introduction Road map 1 ARMA(1,1) model Definition and conditions Moments Estimation 2 ARMA(p,q) model Definition and conditions Moments Estimation 3 Application 4 Appendix FlorianPelgrin (HEC) Univariatetimeseries Sept.2011-Jan.2012 3/32 ARMA(1,1)model Definitionandconditions 1. ARMA(1,1) 1.1. Definition and conditions Definition A stochastic process (Xt)t∈Z is said to be a mixture autoregressive moving average model of order 1, ARMA(1,1), if it satisfies the following equation : X = µ+φX +(cid:15) +θ(cid:15) ∀t t t−1 t t−1 Φ(L)X = µ+Θ(L)(cid:15) t t where θ (cid:54)= 0, θ (cid:54)= 0, µ is a constant term, ((cid:15)t)t∈Z is a weak white noise process with expectation zero and variance σ2 ((cid:15) ∼ WN(0,σ2)), (cid:15) t (cid:15) Φ(L) = 1−φL and Θ(L) = 1+θL. FlorianPelgrin (HEC) Univariatetimeseries Sept.2011-Jan.2012 4/32 ARMA(1,1)model Definitionandconditions Simulation of an ARMA(1,1) with (-0.5;-0.5) Simulation of an ARMA(1,1) with (-0.5;0.5) 4 10 2 5 0 0 -2 -5 -4 -10 0 100 200 300 0 100 200 300 Simulation of an ARMA(1,1) with (0.9;-0.5) Simulation of an ARMA(1,1) with (0.9;0.5) 10 4 5 2 0 0 -5 -2 -10 -4 0 100 200 300 0 100 200 300 FlorianPelgrin (HEC) Univariatetimeseries Sept.2011-Jan.2012 5/32 ARMA(1,1)model Definitionandconditions The properties of an ARMA(1,1) process are a mixture of those of an AR(1) and MA(1) processes : The (stability) stationarity condition is the one of an AR(1) process (or ARMA(1,0) process) : |φ|<1. The invertibility condition is the one of a MA(1) process (or ARMA(0,1) process) : |θ|<1. The representation of an ARMA(1,1) process is fundamental or causal if : |φ|<1 and |θ|<1. The representation of an ARMA(1,1) process is said to be minimal and causal if : |φ|<1,|θ|<1 and φ(cid:54)=θ. FlorianPelgrin (HEC) Univariatetimeseries Sept.2011-Jan.2012 6/32 ARMA(1,1)model Definitionandconditions If (X ) is stable and thus weakly stationary, then (X ) has an infinite t t moving average representation (MA(∞)) : µ X = +(1−φL)−1(1+θL)(cid:15) t t 1−φ ∞ µ (cid:88) = + a (cid:15) k t−k 1−φ k=0 where : a = 1 0 a = φk +θφk−1. k FlorianPelgrin (HEC) Univariatetimeseries Sept.2011-Jan.2012 7/32 ARMA(1,1)model Definitionandconditions If (X ) is invertible, then (X ) has an infinite autoregressive t t representation (AR(∞)) : µ (1−θ∗L)−1(1−φL)X = +(cid:15) t 1−θ∗ t i.e. ∞ µ (cid:88) X = = + b X +(cid:15) t 1−θ∗ k t−k t k=1 where θ∗ = −θ, and : b = −θ∗k −θ∗k−1φ. k FlorianPelgrin (HEC) Univariatetimeseries Sept.2011-Jan.2012 8/32 ARMA(1,1)model Moments 1.2. Moments Definition Let (X ) denote a stationary stochastic process that has a fundamental t ARMA(1,1) representation, X = µ+φX +(cid:15) +θ(cid:15) . Then : t t−1 t t−1 µ E[X ] = ≡ m t 1−φ 1+2φθ+θ2 γ (0) ≡ V(X ) = σ2 X t 1−φ2 (cid:15) (φ+θ)(1+φθ) γ (1) ≡ Cov [X ,X ] = σ2 X t t−1 1−φ2 (cid:15) γ (h) = φγ (h−1) for |h| > 1. X X Proof : See Appendix 1. FlorianPelgrin (HEC) Univariatetimeseries Sept.2011-Jan.2012 9/32 ARMA(1,1)model Moments Definition The autocorrelation function of an ARMA(1,1) process satisfies :  1 if h = 0       ρ (h) = (φ+θ)(1+φθ) if |h| = 1 X 1+2φθ+θ2       φρ (h−1) if |h| > 1. X FlorianPelgrin (HEC) Univariatetimeseries Sept.2011-Jan.2012 10/32

Description:
Lecture 2: ARMA(p,q) models (part 3) Florian Pelgrin University of Lausanne, Ecole des HEC The partial autocorrelation function of an ARMA(1,1) process will
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.