TimeSeriesBook.pdf

# Appendix autoregressive final form definition 121

• 435

This preview shows pages 258–262. Sign up to view the full content.

Appendix: Autoregressive Final Form Definition 12.1 defined the VARMA process { X t } as a solution to the cor- responding multivariate stochastic difference equation (12.1). However, as pointed out by Zellner and Palm (1974) there is an equivalent representation in the form of n univariate ARMA processes, one for each X it . Formally, these representations, also called autoregressive final form or transfer func- tion form (Box and Jenkins, 1976), can be written as det Φ(L) X it = [Φ * (L)Θ(L)] i Z t

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

240 CHAPTER 12. VARMA PROCESSES where the index i indicates the i -th row of Φ * (L)Θ(L). Thereby Φ * (L) denotes the adjugate matrix of Φ(L). 5 Thus each variable in X t may be investigated separately as an univariate ARMA process. Thereby the au- toregressive part will be the same for each variable. Note, however, that the moving-average processes will be correlated across variables. The disadvantage of this approach is that it involves rather long AR and MA lags as will become clear from the following example. 6 Take a simple two-dimensional VAR of order one, i.e. X t = Φ X t - 1 + Z t , Z t WN(0 , Σ). Then the implied univariate processes will be ARMA(2,1) processes. After some straightforward manipulations we obtain: (1 - ( φ 11 + φ 22 )L + ( φ 11 φ 22 - φ 12 φ 21 )L 2 ) X 1 t = Z 1 t - φ 22 Z 1 ,t - 1 + φ 12 Z 2 ,t - 1 , (1 - ( φ 11 + φ 22 )L + ( φ 11 φ 22 - φ 12 φ 21 )L 2 ) X 2 t = φ 21 Z 1 ,t - 1 + Z 2 t - φ 11 Z 2 ,t - 1 . It can be shown by the means given in Sections 1.4.3 and 1.5.1 that the right hand sides are observationally equivalent to MA(1) processes. 5 The elements of the adjugate matrix A * of some matrix A are given by [ A * ] ij = ( - 1) i + j M ij where M ij is the minor (minor determinant) obtained by deleting the i -th column and the j -th row of A (Meyer, 2000, p. 477). 6 The degrees of the AR and the MA polynomial can be as large as np and ( n - 1) p + q , respectively.
Chapter 13 Estimation of Vector Autoregressive Models 13.1 Introduction In this chapter we derive the Least-Squares (LS) estimator for vectorautore- gressive (VAR) models and its asymptotic distribution. For this end, we have to make several assumption which we maintain throughout this chapter. Assumption 13.1. The VAR process { X t } is generated by Φ(L) X t = Z t X t - Φ 1 X t - 1 - · · · - Φ p X t - p = Z t with Z t WN(0 , Σ) , Σ nonsingular, and admits a stationary and causal representation with respect to { Z t } : X t = Z t + Ψ 1 Z t - 1 + Ψ 2 Z t - 2 + . . . = X j =0 Ψ j Z t = Ψ(L) Z t with j =0 k Ψ j k < . Assumption 13.2. The residual process { Z t } is not only white noise, but also independently and identically distributed: Z t IID(0 , Σ) . Assumption 13.3. All fourth moments of Z t exist. In particular, there exists a finite constant c > 0 such that E ( Z it Z jt Z kt Z lt ) c for all i, j, k, l = 1 , 2 , . . . , n, and for all t. 241

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

242 CHAPTER 13. ESTIMATION OF VAR MODELS Note that the moment condition is automatically fulfilled by Gaussian processes. For the ease of exposition, we omit a constant in the VAR. Thus, we consider the demeaned process.
This is the end of the preview. Sign up to access the rest of the document.
• Spring '17
• Raffaelle Giacomini

{[ snackBarMessage ]}

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern