11

Multivariate GARCH Processes

While the volatility of univariate series has been the focus of the previous chapters, modeling the comovements of several series is of great practical importance. When several series displaying temporal or contemporaneous dependencies are available, it is useful to analyze them jointly, by viewing them as the components of a vector-valued (multivariate) process. The standard linear modeling of real time series has a natural multivariate extension through the framework of the vector ARMA (VARMA) models. In particular, the subclass of vector autoregresslve (VAR) models has been widely studied in the econometric literature. This extension entails numerous specific problems and has given rise to new research areas (such as cointegration).

Similarly, it is important to introduce the concept of multivariate GARCH model. For instance, asset pricing and risk management crucially depend on the conditional covariance structure of the assets of a portfolio. Unlike the ARMA models, however, the GARCH model specification does not suggest a natural extension to the multivariate framework. Indeed, the (conditional) expectation of a vector of size m is a vector of size m, but the (conditional) variance is an m × m matrix. A general extension of the univariate GARCH processes would involve specifying each of the m(m + l)/2 entries of this matrix as a function of its past values and the past values of the other entries. Given the excessive number of parameters that this approach would entail, it is not feasible from a statistical point of view. An alternative approach is to Introduce some specification constraints which, while preserving a certain generality, make these models operational.

We start by reviewing the main concepts for the analysis of the multivariate time series.

11.1 Multivariate Stationary Processes

In this section, we consider a vector process (Xt)t2208_fmt2124_fmt of dimension m, Xt = (X1t,…, Xmt)'′. The definition of strict stationarity (see Chapter 1, Definition 1.1) remains valid for vector processes, while second-order stationarity is defined as follows.

Definition 11.1 (Second-order stationarity) The process (Xt) is said to be second-order stationary if:

(i) c11-ie273001_fmt < ∞ 2200_fmtt 2208_fmt 2124_fmt, i = 1,…, m;

(ii) EXt = μ, 2200_fmtt 2208_fmt 2124_fmt;

(iii) Cov(Xt, Xt+h) = E[(Xt − μ)(Xt+h − μ)′] = 0413_fmt(h), 2200_fmtt, h 2208_fmt 2124_fmt.

The function 0413_fmt(·), taking values in the space of m × m matrices, is called the autocovariance function of (Xt).

Obviously 0413_fmtX(h) = 0413_fmtX(−h)′. In particular, 0413_fmtX(0) = Var(Xt) is a symmetric matrix.

The simplest example of a multivariate stationary process is white noise, defined as a sequence of centered and uncorrelated variables whose covariance matrix is time-independent.

The following property can be used to construct stationary processes by linear transformation of another stationary process.

Theorem 11.1 (Stationary linear filter) Let (Zt) denote a stationary process, Zt 2208_fmt 211D_fmtm. Let (Ck) k2208_fmt2124_fmt denote a sequence of nonrandom n × m matrices, such that, for all i = 1, …, n,for all j = 1,… ,m, c11-ie274001_fmt, where c11-ie274002_fmt. Then the 211D_fmtn-valued process defined by c11-ie274003_fmt is stationary and we have, in obvious notation,

c11ue001_fmt

The proof of an analogous result is given by Brockwell and Davis (1991, pp. 83–84) and the arguments used extend straightforwardly to the multivariate setting. When, in this theorem, (Zt) is a white noise and Ck = 0 for all k < 0, (Xt) is called a vector moving average process of infinite order, VMA(∞). A multivariate extension of Wold’s representation theorem (see Hannan, 1970, pp. 157–158) states that if (Xt) is a stationary and purely nondeterministic process, it can be represented as an infinite-order moving average,

(11.1) c11e001_fmt

where (2208_fmtt) is an (m × 1) white noise, B is the lag operator, c11-ie274004_fmt, and the matrices Ck are not necessarily absolutely summable but satisfy the (weaker) condition c11-ie274005_fmt for any matrix norm 2016_fmt · 2016_fmt. The following definition generalizes the notion of a scalar ARMA process to the multivariate case.

Definition 11.2 (VARMA(p, q) process) An 211D_fmtm-valued process (Xt)t2208_fmt2124_fmt is called a vector ARMA process of orders p and q (VARMA(p, q)) if (Xt)t2208_fmt2124_fmt is a stationary solution to the difference equation

(11.2) c11e002_fmt

where (2208_fmtt) is an (m × 1) white noise with covariance matrix 2126_fmt, c is an m × 1 vector, and 0424_fmt(z) = Im − Φ1z − … − c11-ie274007_fmt and Ψ(z) = Im − Ψ1z − … − c11-ie274008_fmt are matrix-valued polynomials.

Denote by det(A), or more simply |A| when there is no ambiguity, the determinant of a square matrix A. A sufficient condition for the existence of a stationary and invertible solution to the preceding equation is

c11ue002_fmt

(see Brockwell and Davis, 1991, Theorems 11.3.1 and 11.3.2).

When p = 0, the process is called vector moving average of order q (VMA(q)); when q = 0, the process is called vector autoregressive of order p (VAR(p)).

Note that the determinant |Φ(z)| is a polynomial admitting a finite number of roots z1, …, zmp. Let δ = mini |zi| > 1. The power series expansion

(11.3) c11e003_fmt

where A* denotes the adjoint of the matrix A (that is, the transpose of the matrix of the cofactors of A), is well defined for |z| < δ, and is such that Φ(z)−1Φ(z) = I. The matrices Ck are recursively obtained by

(11.4) c11e004_fmt

11.2 Multivariate GARCH Models

As in the univariate case, we can define multivariate GARCH models by specifying their first two conditional moments. An 211D_fmtm-valued GARCH process (2208_fmtt), with 2208_fmtt = (2208_fmt1t, …, 2208_fmtmt)′, must then satisfy, for all t 2208_fmt 2124_fmt,

(11.5) c11e005_fmt

The multivariate extension of the notion of the strong GARCH process is based on an equation of the form

(11.6) c11e006_fmt

where (ηt) is a sequence of iid 211D_fmtm-valued variables with zero mean and identity covariance matrix. The matrix c11-ie275002_fmt can be chosen to be symmetric and positive definite1 but it can also be chosen to be triangular, with positive diagonal elements (see, for instance, Harville, 1997, Theorem 14.5.11). The latter choice may be of interest because if, for instance, c11-ie275002_fmt is chosen to be lower triangular, the first component of 2208_fmtt only depends on the first component of ηt. When m = 2, we can thus set

(11.7) c11e007_fmt

where ηit and hij,t denote the generic elements of ηt and Ht.

Note that any square integrable solution (2208_fmtt) of (11.6) is a martingale difference satisfying (11.5).

Choosing a specification for Ht is obviously more delicate than in the univariate framework because: (i) Ht should be (almost surely) symmetric, and positive definite for all t; (ii) the specification should be simple enough to be amenable to probabilistic study (existence of solutions, stationarity, …), while being of sufficient generality; (iii) the specification should be parsimonious enough to enable feasible estimation. However, the model should not be too simple to be able to capture the - possibly sophisticated - dynamics in the covariance structure.

Moreover, it may be useful to have the so-called stability by aggregation property. If 2208_fmtt satisfies (11.5), the process (c11-ie275003_fmt) defined by c11-ie275003_fmt = P2208_fmtt, where P is an invertible square matrix, is such that

(11.8)c11e008_fmt

The stability by aggregation of a class of specifications for Ht requires that the conditional variance matrices c11-ie276001_fmt belong to the same class for any choice of P. This property is particularly relevant in finance because if the components of the vector 2208_fmtt are asset returns, c11-ie275003_fmt is a vector of portfolios of the same assets, each of its components consisting of amounts (coefficients of the corresponding row of P) of the initial assets.

11.2.1 Diagonal Model

A popular specification, known as the diagonal representation, is obtained by assuming that each element hk2113_fmt,t of the covariance matrix ht is formulated in terms only of the product of the prior k and 2113_fmt returns. Specifically,

c11ue003_fmt

with ωk2113_fmt = ω2113_fmtk, c11-ie276002_fmt, c11-ie276003_fmt for all (k,2113_fmt). For m = 1 this model coincides with the usual univariate formulation. When m > 1 the model obviously has a large number of parameters and will not in general produce positive definite covariance matrices Ht. We have

c11ue004_fmt

where 2299_fmt denotes the Hadamard product, that is, the element by element product.2 Thus, in the ARCH case (p = 0), sufficient positivity conditions are that 2126_fmt is positive definite and the A(i) are positive semi-definite, but these constraints do not easily generalize to the GARCH case. We shall give further positivity conditions obtained by expressing the model in a different way, viewing it as a particular case of a more general class.

It is easy to see that the model is not stable by aggregation: for instance, the conditional variance of 2208_fmt1,t + 2208_fmt2,t can in general be expressed as a function of the c11-ie276005_fmt and c11-ie276006_fmt but not of the (2208_fmt1,t−i + 2208_fmt2,t−i)2. A final drawback of this model is that there is no interaction between the different components of the conditional covariance, which appears unrealistic for applications to financial series.

In what follows we present the main specifications introduced in the literature, before turning to the existence of solutions. Let η denote a probability distribution on 211D_fmtm, with zero mean and unit covariance matrix.

11.2.2 Vector GARCH Model

The vector GARCH (VEC-GARCH) model is the most direct generalization of univariate GARCH: every conditional covariance is a function of lagged conditional variances as well as lagged cross-products of all components. In some sense, everything is explained by everything, which makes this model very general but also not very parsimonious.

Denote by vech(·) the operator that stacks the columns of the lower triangular part of its argument square matrix (if A = (aij), then vech(A) = (a11, a21, …, am1, a22, …, am2, …, amm)′). The next definition is a natural extension of the standard GARCH(p, q) specification.

Definition 11.3 (VEC-GARCH(p, q) process) Lett) be a sequence of iid variables with distribution η. The process (2208_fmtt) is said to admit a VEC-GARCH(p, q) representation (relative to the sequencet)) if it satisfies

(11.9) c11e009_fmt

where ω is a vector of size {m(m + l)/2} × 1, and the A(i) and B(j) are matrices of dimension m(m + l)/2 × m(m + l)/2.

Remark 11.1 (The diagonal model is a special case of the VEC-GARCH model) The diagonal model admits a vector representation, obtained for diagonal matrices A(i) and B(j).

We will show that the class of VEC-GARCH models is stable by aggregation. Recall that the vec(·) operator converts any matrix to a vector by stacking all the columns of the matrix into one vector. It is related to the vech operator by the formulas

(11.10) c11e010_fmt

where A is any m × m symmetric matrix, Dm is a full-rank m2 × m(m + l)/2 matrix (the so-called ‘duplication matrix’), whose entries are only 0 and 1, D+m_fmt = (D′mDm)−1D′m.3 We also have the relation

(11.11) c11e011_fmt

where 2297_fmt denotes the Kronecker matrix product,4 provided the product ABC is well defined.

Theorem 11.2 (The VEC-GARCH is stable by aggregation) Let (2208_fmtt) be a VEC-GARCH(p, q) process. Then, for any invertible m × m matrix P, the process c11-ie275003_fmt = P2208_fmtt is a VEC-GARCH(p, q) process.

Proof. Setting c11-ie278001_fmt = PHtP′, we have 2208_fmtt = c11-ie278002_fmtηt and

c11ue006_fmt

where

c11ue007_fmt

To derive the form of c11-ie278005_fmt we use

c11ue008_fmt

and for c11-ie278007_fmt we use

c11ue009_fmt      Box_fmt

Positivity Conditions

We now seek conditions ensuring the positivity of Ht. A generic element of

c11ue010_fmt

is denoted by hk2113_fmt,t (k2113_fmt) and we will denote by c11-ie278011_fmt c11-ie278012_fmt the entry of A(i) (B(j)) located on the same row as hk2113_fmt,t and belonging to the same column as the element hk′2113_fmt′,t of ht. We thus have an expression of the form

c11ue011_fmt

Denoting by c11-ie278013_fmt the m × m symmetric matrix with (k2113_fmt′)th entries c11-ie278011_fmt/2, for k′ ≠ 2113_fmt′ and the elements c11-ie278011_fmt on the diagonal, the preceding equality is written as

(11.12) c11e012_fmt

In order to obtain a more compact form for the last part of this expression, let us introduce the spectral decomposition of the symmetric matrices Ht, assumed to be positive semi-definite. We have c11-ie278014_fmt where c11-ie278015_fmt is an orthogonal matrix of eigenvectors c11-ie278016_fmt associated with the (positive) eigenvalues c11-ie278017_fmt of Ht. Defining the matrices c11-ie278018_fmt by analogy with the c11-ie278019_fmt, we get

(11.13) c11e013_fmt

Finally, consider the m2 × m2 matrix admitting the block form c11-ie279001_fmt, and let c11-ie279002_fmt. The preceding expressions are equivalent to

(11.14) c11e014_fmt

where 2126_fmt is the symmetric matrix such that vech(2126_fmt) = ω.

In this form, it is evident that the assumption

(11.15) c11e015_fmt

ensures that if the Ht−j are almost surely positive definite, then so is Ht.

Example 11.1 (Three representations of a vector ARCH(l) model) For p = 0, q = 1 and m = 2, the conditional variance is written, in the form (11.9), as

c11ue012_fmt

in the form (11.12) as

c11ue013_fmt

and in the form (11.14) as

c11ue014_fmt

This example shows that, even for small orders, the VEC model potentially has an enormous number of parameters, which can make estimation of the parameters computationally demanding. Moreover, the positivity conditions are not directly obtained from (11.9) but from (11.14), involving the spectral decomposition of the matrices Ht−j.

The following classes provide more parsimonious and tractable models.

11.2.3 Constant Conditional Correlations Models

Suppose that, for a multivariate GARCH process of the form (11.6), all the past information on 2208_fmtkt, involving all the variables 2208_fmt2113_fmt,t−i, is summarized in the variable hkk,t, with Ehkk,t = c11-ie279003_fmt Then, letting c11-ie279004_fmt, we define for all k a sequence of iid variables with zero mean and unit variance. The variables c11-ie280001_fmt are generally correlated, so let c11-ie280002_fmt, where c11-ie280003_fmt . The conditional variance of

c11ue015_fmt

is then written as

(11.16) c11e016_fmt

By construction, the conditional correlations between the components of 2208_fmtt are time-invariant:

c11ue016_fmt

To complete the specification, the dynamics of the conditional variances hkk,t has to be defined. The simplest constant conditional correlations (CCC) model relies on the following univariate GARCH specifications:

(11.17) c11e017_fmt

where ωk > 0, ak,i ≥ 0, bk,j ≥ 0, − 1 ≤ ρk2113_fmt ≤ 1, ρkk = 1, and R is symmetric and positive semi-definite. Observe that the conditional variances are specified as in the diagonal model. The conditional covariances clearly are not linear in the squares and cross products of the returns.

In a multivariate framework, it seems natural to extend the specification (11.17) by allowing hkk,t to depend not only on its own past, but also on the past of all the variables 2208_fmt2113_fmt,t. Set

c11ue017_fmt

Definition 11.4 (CCC-GARCH(p, q) process) Lett) be a sequence of iid variables with distribution η. A process (2208_fmtt) is called CCC-GARCH(p, q) if it satisfies

(11.18) c11e018_fmt

where R is a correlation matrix, omega-bar_fmt a m × 1 vector with positive coefficients, and the Ai; and Bj are m × m matrices with nonnegative coefficients.

We have c11-ie280004_fmt, where c11-ie280005_fmt = R½ηt is a centered vector with covariance matrix R. The components of 2208_fmtt thus have the usual expression, c11-ie280006_fmt, but the conditional variance hkk,t depends on the past of all the components of 2208_fmtt.

Note that the conditional covariances are generally nonlinear functions of the components of 2208_fmtt−i2208_fmtt−i and of past values of the components of Ht. Model (11.18) is thus not a VEC-GARCH model, defined by (11.9), except when R is the identity matrix.

One advantage of this specification is that a simple condition ensuring the positive definiteness of Ht is obtained through the positive coefficients for the matrices Ai and Bj and the choice of a positive definite matrix for R. We shall also see that the study of the stationarity is remarkably simple.

Two limitations of the CCC model are, however, (i) its nonstability by aggregation and (ii) the arbitrary nature of the assumption of constant conditional correlations.

11.2.4 Dynamic Conditional Correlations Models

Dynamic conditional correlations GARCH (DCC-GARCH) models are an extension of CCC-GARCH, obtained by introducing a dynamic for the conditional correlation. Hence, the constant matrix R in Definition 11.4 is replaced by a matrix Rt which is measurable with respect to the past variables {2208_fmtu, u < t}. For reasons of parsimony, it seems reasonable to choose diagonal matrices Ai and Bi in (11.18), corresponding to univariate GARCH models for each component as in (11.17). Different DCC models are obtained depending on the specification of Rt. A simple formulation is

(11.19) c11e019_fmt

where the θi are positive weights summing to 1, R is a constant correlation matrix, and Ψt−1 is the empirical correlation matrix of 2208_fmtt−1,…, 2208_fmtt−M. The matrix Rt is thus a correlation matrix (see Exercise 11.9). Equation (11.19) is reminiscent of the GARCH(1, 1) specification, θ1R playing the role of the parameter ω, θ2 that of α, and θ3 that of β.

Another way of specifying the dynamics of Rt is by setting

c11ue018_fmt

where diag Qt is the diagonal matrix constructed with the diagonal elements of Qt, and Qt is a sequence of covariance matrices which is measurable with respect to σ (2208_fmtu, u < t). A natural parameterization is

(11.20) c11e020_fmt

where Q is a covariance matrix. Again, the formulation recalls the GARCH(1, 1) model. Though different, both specifications (11.19) and (11.20) allow us to test the assumption of constant conditional covariance matrix, by considering the restriction θ2 = θ3 = 0. Note that the same θ2 and θ3 coefficients appear in the different conditional correlations, which thus have very similar dynamics. The matrices R and Q are often estimated/replaced by the empirical correlation and covariance matrices. In this approach a DCC model of the form (11.19) or (11.20) thus introduces only two more parameters than the CCC formulation.

11.2.5 BEKK-GARCH Model

The BEKK acronym refers to a specific parameterization of the multivariate GARCH model developed by Baba, Engle, Kraft and Kroner, in a preliminary version of Engle and Kroner (1995).

Definition 11.5 (BEKK-GARCH(p, q) process) Lett) denote an iid sequence with common distribution η. The process (2208_fmtt) is called a strong GARCH(p, q), with respect to the sequence (2208_fmtt), if it satisfies

c11ue019_fmt

where K is an integer, 2126_fmt, Aik and Bjk are square m × m matrices, and 2126_fmt is positive definite.

The specification obviously ensures that if the matrices Ht−i, i = 1, …, p, are almost surely positive definite, then so is Ht.

To compare this model with the representation (11.9), let us derive the vector form of the equation for Ht. Using the relations (11.10) and (11.11), we get

c11ue020_fmt

The model can thus be written in the form (11.9), with

(11.21) c11e021_fmt

for i = 1,…, q and j = 1,…, p. In particular, it can be seen that the number of coefficients of a matrix A(i) in (11.9) is [m(m + l)/2]2, whereas it is Km2 in this particular case.

The BEKK class contains (Exercise 11.13) the diagonal models obtained by choosing diagonal matrices Aik and Bjk. The following theorem establishes a converse to this property.

Theorem 11.3 For the model defined by the diagonal vector representation (11.9) with

c11ue021_fmt

where c11-ie282001_fmt and c11-ie282002_fmt are m × m symmetric positive semi-definite matrices, there exist matrices Aik and Bjk such that (11.21) holds, for K = m.

Proof. There exists an upper triangular matrix

c11ue022_fmt

such that c11-ie283001_fmt. Let c11-ie284001_fmt where r = m − k + 1, for k = 1.,…, m. It is easy to show that the first equality in (11.21) is satisfied with K = m. The second equality is obtained similarly.       Box_fmt

Example 11.2 By way of illustration, consider the particular case where m = 2, q = K = 1 and p = 0. If A = (aij) is a 2 × 2 matrix, it is easy to see that

c11ue023_fmt

Hence, canceling out the unnecessary indices,

c11ue024_fmt

In particular, the diagonal models belonging to this class are of the form

c11ue025_fmt

Remark 11.2 (Interpretation of the BEKK coefficients) Example 11.2 shows that the BEKK specification imposes highly artificial constraints on the volatilities and covolatilities of the components. As a consequence, the coefficients of a BEKK representation are difficult to interpret.

Remark 11.3 (Identifiability) Identifiability of a BEKK representation requires additional constraints. Indeed, the same representation holds if Aik is replaced by −Aik, or if the matrices A1k,…, Aqk and A1k,…, Aqk are permuted for kk′.

Example 11.3 (A general and identifiable BEKK representation) Consider the case m = 2, q = 1 and p = 0. Suppose that the distribution η is nondegenerate, so that there exists no nontrivial constant linear combination of a finite number of the 2208_fmtk,t−i2208_fmt2113_fmt,t−i Let

c11ue026_fmt

where 2126_fmt is a symetric positive definite matrix,

c11ue027_fmt

with a11,1 ≥ 0, a12,3 ≥ 0, a21,2 ≥ 0 and a241,4 ≥ 0.

Let us show that this BEKK representation is both identifiable and quite general. Easy, but tedious, computation shows that an expression of the form (11.9) holds with

c11ue028_fmt

In view of the sign constraint, the (1, l)th element of A(1) allows us to identify a11,1. The (1, 2)th and (2, l)th elements then allow us to find a12,1 and a21,1, whence the (2, 2)th element yields a22,1. The two elements of A3 are deduced from the (1, 3)th and (2, 3)th elements of A(1) and from the constraint a12,3 ≥ 0 (which could be replaced by a constraint on the sign of a22,3). A2 is identified similarly, and the nonzero element of A4 is finally identified by considering the (3, 3)th element of A(1).

In this example, the BEKK representation contains the same number of parameters as the corresponding VEC representation, but has the advantage of automatically providing a positive definite solution Ht.

It is interesting to consider the stability by aggregation of the BEKK class.

Theorem 11.4 (Stability of the BEKK model by aggregation) Let (2208_fmtt) be a BEKK-GARCH (p,q) process. Then, for any invertible m × m matrix P, the process c11-ie275003_fmt = P2208_fmtt is a BEKK-GARCH (p, q) process.

Proof. Letting c11-ie284003_fmt = PHtP′, c11-ie284004_fmt = P2126_fmtP′,c11-ie284005_fmt = PAikP−1 and c11-ie284005_fmt = PBjkP−1 we get

c11ue029_fmt

and, c11-ie284004_fmt being a positive definite matrix, the result is proved.       Box_fmt

As in the univariate case, the ‘square’ of the (2208_fmtt) process is the solution of an ARMA model. Indeed, define the innovation of the process vec(2208_fmtt2208_fmtt):

(11.22) c11e022_fmt

Applying the vec operator, and substituting the variables vec(Ht−j) in the model of Definition 11.5 by c11-ie284007_fmt, we get the representation

(11.23) c11e023_fmt

where r = max(p, q), with the convention Aik = 0 (Bjk = 0) If i > q (j > p). This representation cannot be used to obtain stationarity conditions because the process (vt) is not iid in general. However, it can be used to derive the second-order moment, when it exists, of the process 2208_fmtt as

c11ue030_fmt

that is,

c11ue031_fmt

provided that the matrix in braces is nonsingular.

11.2.6 Factor GARCH Models

In these models, it is assumed that a nonsingular linear combination ft of the m components of 2208_fmtt, or an exogenous variable summarizing the comovements of the components, has a GARCH structure.

Factor models with idiosyncratic noise

A very popular factor model links individual returns 2208_fmtit to the market return ft through a regression model

(11.24) c11e024_fmt

The parameter βi can be Interpreted as a sensitivity to the factor, and the noise ηit as a specific risk (often called idiosyncratic risk) which is conditionally uncorrelated with ft. It follows that Ht = 2126_fmt + λtββ′ whereβ is the vector of sensitivities, λt is the conditional variance of ft and 2126_fmt is the covariance matrix of the idiosyncratic terms. More generally, assuming the existence of r conditionally uncorrelated factors, we obtain the decomposition

(11.25) c11e025_fmt

It is not restrictive to assume that the factors are linear combinations of the components of 2208_fmtt (Exercise 11.10). If, in addition, the conditional variances λjt are specified as univariate GARCH, the model remains parsimonious in terms of unknown parameters and (11.25) reduces to a particular BEKK model (Exercise 11.11). If 2126_fmt is chosen to be positive definite and if the univariate series (λjt)t, j = 1,…, r, are independent, strictly and second-order stationary, then it is clear that (11.25) defines a sequence of positive definite matrices (Ht) that are strictly and second-order stationary.

Principal components GARCH model

The concept of factor is central to principal components analysis (PCA) and to other methods of exploratory data analysis. PCA relies on decomposing the covariance matrix V of m quantitative variables as V = PΛP′ where Λ is a diagonal matrix whose elements are the eigenvalues λ1 ≥ λ2 ≥ … ≥ λm of V, and where P is the orthonormal matrix of the corresponding eigenvectors. The first principal component is the linear combination of the m variables, with weights given by the first column of P, which, in some sense, is the factor which best summarizes the set of m variables (Exercise 11.12). There exist m principal components, which are uncorrelated and whose variances λ1,…, λm (and hence whose explanatory powers) are in decreasing order. It is natural considering this method for extracting the key factors of the volatilities of the m components of 2208_fmtt.

We obtain a principal component GARCH (PC-GARCH) or orthogonal GARCH (O-GARCH) model by assuming that

(11.26) c11e026_fmt

where P is an orthogonal matrix (P′ = P−1) and Λt = diag (λ1t,…, λmt), where the λit are the volatilities, which can be obtained from univariate GARCH-type models. This is equivalent to assuming

(11.27) c11e027_fmt

where ft = P2208_fmtit is the principal component vector, whose components are orthogonal factors. If univariate GARCH(1, 1) models are used for the factors c11-ie285001_fmt then

(11.28) c11e028_fmt

Remark 11.4 (Interpretation, factor estimation and extensions)

1. Model (11.26) can also be interpreted as a full-factor GARCH (FF-GARCH) model, that is, a model with as many factors as components and no idiosyncratic term. Let P(·, j) be the jth column of P (an eigenvector of Ht associated with the eigenvalue λjt). We get a spectral expression for the conditional variance,

c11ue032_fmt

which is of the form (11.25) with an idiosyncratic variance 2126_fmt = 0.

2. A PCA of the conditional variance Ht should, in full generality, give Ht = PtΛtPt with factors (that is, principal components) ft = Pt2208_fmtt. Model (11.26) thus assumes that all factors are linear combinations, with fixed coefficients, of the same returns 2208_fmtit. For instance, the first

factor f1t is the conditionally most risky factor (with the largest conditional variance λ1t, see Exercise 11.12). But since it is assumed that the direction of f1t is fixed, in the subspace of 211D_fmtm generated by the components of 2208_fmtit, the first factor is also the most risky unconditionally. This can be seen through the PCA of the unconditional variance H = EHt = PΛP′, which is assumed to exist.

3. It is easy to estimate P by applying PCA to the empirical variance c11-ie286001a_fmt, where c11-ie286001_fmt. The components of c11-ie286002_fmt are specified as GARCH-type univariate models. Estimation of the conditional variance c11-ie286003_fmt thus reduces to estimating m univariate models.

4. It is common practice to apply PCA on centered and standardized data, in order to remove the influence of the units of the various variables. For returns 2208_fmtit, standardization does not seem appropriate if one wishes to retain a size effect, that is, if one expects an asset with a relatively large variance to have more weight in the riskier factors.

5. In the spirit of the standard PCA, it is possible to only consider the first r principal components, which are the key factors of the system. The variance Ht is thus approximated by

(11.29) c11e029_fmt

where the c11-ie286004_fmt are estimated from simple univariate models, such as GARCH(1, 1) models of the form (11.28), the matrix P-carret_fmt is obtained from PCA of the empirical covariance matrix c11-ie286005_fmt, and the factors are approximated by c11-ie286006_fmt. Instead of the approximation (11.29), one can use

(11.30) c11e030_fmt

The approximation in (11.30) is as simple as (11.29) and does not require additional computations (in particular, the r GARCH equations are retained) but has the advantage of providing an almost surely invertible estimation of Ht (for fixed n), which is required in the computation of certain statistics (such as the AIC-type information criteria based on the Gaussian log-likelihood).

6. Note that the assumption that P is orthogonal can be restrictive. The class of generalized orthogonal GARCH (GO-GARCH) processes assumes only that P is any nonsingular matrix.

11.3 Stationarity

In this section, we will first discuss the difficulty of establishing stationarity conditions, or the existence of moments, for multivariate GARCH models. For the general vector model (11.9), and in particular for the BEKK model, there exist sufficient stationarity conditions. The stationary solution being nonexplicit, we propose an algorithm that converges, under certain assumptions, to the stationary solution. We will then see that the problem is much simpler for the CCC model (11.18).

11.3.1 Stationarity of VEC and BEKK Models

It is not possible to provide stationary solutions, in explicit form, for the general VEC model (11.9). To illustrate the difficulty, recall that a univariate ARCH(l) model admits a solution 2208_fmtt = σtηt with σt explicitly given as a function of {ηt−u, u > 0} as the square root of

c11ue033_fmt

provided that the series converges almost surely. Now consider a bivariate model of the form (11.6) with Ht = I2 + α2208_fmtt−12208_fmtt−1 where α is assumed, for the sake of simplicity, to be scalar and positive. Also choose c11-ie275002_fmt to be lower triangular so as to have (11.7). Then

c11ue034_fmt

It can be seen that, given ηt−1, the relationship between h11,t and h11,t−1 is linear, and can be iterated to yield

c11ue035_fmt

under the constraint c11-ie287003_fmt. In contrast, the relationships between h12,t, or h22,t, and the components of Ht−1 are not linear, which makes it impossible to express h12,t and h22,t as a simple function of α, {ηt−1, ηt−2, …, ηt−k} and Ht−k for k ≥ 1. This constitutes a major obstacle for determining sufficient stationarity conditions.

Remark 11.5 (Stationarity does not follow from the ARMA model) Similar to (11.22), letting c11-ie287004_fmt, we obtain the ARMA representation

c11ue036_fmt

by setting C(i) = A(i) + B(i) and by using the usual notation and conventions. In the literature, one may encounter the argument that the model is weakly stationary if the polynomial z 21A6_fmt det(c11-ie287005_fmt) has all its roots outside the unit circle (s = m(m + l)/2). Although the result is certainly true with additional assumptions on the noise density (see Theorem 11.5 and the subsequent discussion), the argument is not correct since

c11ue037_fmt

constitutes a solution only if vt = vech (2208_fmtt2208_fmtt) – vech(Ht) can be expressed as a function of {ηt−u,u > 0}.

Boussama (2006) obtained the following stationarity condition. Recall that ρ(A) denotes the spectral radius of a square matrix A.

Theorem 11.5 (Stationarity and ergodicity) There exists a strictly stationary and nonanticipative solution of the vector GARCH model (11.9), if:

(i) the positivity condition (11.15) is satisfied;

(ii) the distribution of η has a density, positive on a neighborhood of 0, with respect to the Lebesgue measure on 211D_fmtm;

(iii) ρs (c11-ie287006_fmt) > 1.

This solution is unique, β-mixing and ergodic.

In the particular case of the BEKK model (11.21), condition (III) takes the form

c11ue038_fmt

The proof of Theorem 11.5 relies on sophisticated algebraic tools. Assumption (ii) is a standard technical condition for showing the β-mixing property (but is of no use for stationarity). Note that condition (iii), written as c11-ie288001_fmt < 1 in the univariate case, is generally not necessary for the strict stationarity.

This theorem does not provide explicit stationary solutions, that is, a relationship between 2208_fmtt and the ηt−i. However, it is possible to construct an algorithm which, when it converges, allows a stationary solution to the vector GARCH model (11.9) to be defined.

Construction of a stationary solution

For any t, k 2208_fmt 2124_fmt, we define

c11ue039_fmt

and, recursively on k ≥ 0,

(11.31) c11e031_fmt

with c11-ie288002_fmt = c11-ie288003_fmt.

Observe that, for k ≥ 1,

c11ue040_fmt

where fk is a measurable function and H(k) is a square matrix. c11-ie288004_fmt and c11-ie288005_fmt are thus stationary processes whose components take values in the Banach space L2 of the (equivalence classes of) square integrable random variables. It is then clear that (11.9) admits a strictly stationary solution, which is nonanticipative and ergodic, if, for all t,

(11.32) c11e032_fmt

Indeed, letting c11-ie288006_fmt and c11-ie288007_fmt, and taking the limit of each side of (11.31), we note that (11.9) is satisfied. Moreover, (2208_fmtt) constitutes a strictly stationary and nonanticipative solution, because 2208_fmtt is a measurable function of {ηu, ut}. In view of Theorem A.1, such a process is also ergodic. Note also that if Ht exists, it is symmetric and positive definite because the matrices c11-ie288009_fmt are symmetric and satisfy

c11ue041_fmt

This solution (2208_fmtt) is also second-order stationary if

(11.33) c11e033_fmt

Let

c11ue042_fmt

From Exercise 11.8 and its proof, we obtain (11.32), and hence the existence of strictly stationary solution to the vector GARCH equation (11.9), if there exists ρ 2208_fmt]0, 1[ such that c11-ie288009_fmt almost surely as k → ∞, which is equivalent to

(11.34) c11e034_fmt

Similarly, we obtain (11.33) if c11-ie289001_fmt. The criterion in (11.34) is not very explicit but the left-hand side of the inequality can be evaluated by simulation, just as for a Lyapunov coefficient.

11.3.2 Stationarity of the CCC Model

In model (11.18), letting c11-ie289002_fmt, we get

c11ue043_fmt

Multiplying by Υt the equation for h_underline_t_fmt, we thus have

c11ue044_fmt

which can be written

(11.35) c11e035_fmt

where

c11ue045_fmt

and

(11.36) c11e036_fmt

is a (p + q)m × (p + q)m matrix.

We obtain a vector representation, analogous to (2.16) obtained in the univariate case. This allows to state the following result.

Theorem 11.6 (Strict stationarity of the CCC model) A necessary and sufficient condition for the existence of a strictly stationary and nonanticipative solution process for model (11.18) is γ < 0, where y is the top Lyapunov exponent of the sequence {At, t 2208_fmt 2124_fmt} defined in (11.36). This stationary and nonanticipative solution, when γ < 0, is unique and ergodic.

Proof. The proof is similar to that of Theorem 2.4. The variables ηt admitting a variance, the condition E log+ 2016_fmtAt2016_fmt < ∞ is satisfied.

It follows that when γ < 0, the series

(11.37) c11e037_fmt

converges almost surely for all t. A strictly stationary solution to model (11.18) is obtained as c11-ie290002_fmt where c11-ie290003_fmt denotes the (q + l)th subvector of size m of c11-ie290004_fmt This solution is thus nonanticipative and ergodic. The proof of the uniqueness is exactly the same as in the univariate case.

The proof of the necessary part can also be easily adapted. From Lemma 2.1, it is sufficient to prove that limt→∞ 2016_fmt A0At 2016_fmt = 0. It suffices to show that, for 1 ≤ ip + q,

(11.38) c11e038_fmt

where c11-ie290005_fmt and ei is the ith element of the canonical basis of 211D_fmtp+q, since any vector x of 211D_fmtm(p + q) can be uniquely decomposed as c11-ie290006_fmt where xi 2208_fmt 211D_fmtm. As in the univariate case, the existence of a strictly stationary solution implies that A0 … A−kunderline-b_fmt−k−1 tends to 0, almost surely, as k → ∞. It follows that, using the relation c11-ie290007_fmt, we have

(11.39) c11e039_fmt

Since the components of omega-underline_fmt are strictly positive, (11.38) thus holds for i = q + 1. Using

(11.40) c11e040_fmt

with the convention that c11-ie290008_fmt, for i = 1 we obtain

c11ue046_fmt

where the inequalities are taken componentwise. Therefore, (11.38) holds true for i = q + 2, and by induction, for i = q + j, j = 1,…,p in view of (11.40). Moreover, since c11-ie290010_fmt, (11.38) holds for i = q. We reach the same conclusion for the other values of i using an ascending recursion, as in the univariate case.       Box_fmt

The following result provides a necessary strict stationarity condition which is simple to check.

Corollary 11.1 (Consequence of the strict stationarity) Let γ denote the top Lyapunov exponent of the sequence {At,t 2208_fmt 2124_fmt} defined in (11.36). Consider the matrix polynomial defined by: 212C_fmt(z) = ImzB1 − … − zpBp, z 2208_fmt 2102_fmt. Let

c11ue047_fmt

Then, if γ< 0 the following equivalent properties hold:

1. The roots of det 212C_fmt (z) are outside the unit disk.

2. ρ(double-struck-B_fmt) < 1.

Proof. Because all the entries of the matrices At are positive, it is clear that γ is larger than the top Lyapunov exponent of the sequence (A*t) obtained by replacing the matrices Ai: by 0 in At. It is easily seen that the top Lyapunov coefficient of (A*t) coincides with that of the constant sequence equal to double-struck-B_fmt, that is, with ρ(double-struck-B_fmt). It follows that γ ≥ log ρ(double-struck-B_fmt). Hence γ < 0 entails that all the eigenvalues of double-struck-B_fmt are outside the unit disk. Finally, in view of Exercise 11.14, the equivalence between the two properties follows from

c11ue048_fmt      ◀Box_fmt

Corollary 11.2 Suppose that γ < 0. Let 2208_fmtt be the strictly stationary and nonanticipative solution of model (11.18). There exists s < 0 such that c11-ie291002_fmt and c11-ie291003_fmt.

Proof. It is shown in the proof of Corollary 2.3 that the strictly stationary solution defined by (11.37) satisfies c11-ie291004_fmt for some s < 0. The conclusion follows from c11-ie291005_fmt and c11-ie291006_fmt.       Box_fmt

11.4 Estimation of the CCC Model

We now turn to the estimation of the m-dimensional CCC-GARCH(p, q) model by the quasi-maximum likelihood method. Recall that (2208_fmtt) is called a CCC-GARCH(p, q) if it satisfies

(11.41) c11e041_fmt

where R is a correlation matrix, omega-underline_fmt is a vector of size m × 1 with strictly positive coefficients, the Ai and Bj are matrices of size m × m with positive coefficients, and (ηt) is an iid sequence of centered variables in 211D_fmtm with identity covariance matrix.

As in the univariate case, the criterion is written as if the iid process were Gaussian.

The parameters are the coefficients of the matrices omega-underline_fmt, Ai and Bj, and the coefficients of the lower triangular part (excluding the diagonal) of the correlation matrix R = (ρij). The number of unknown parameters is thus

c11ue049_fmt

The parameter vector is denoted by

c11ue050_fmt

where ρ′ = (ρ21 …, ρm1 ρ32,…, ρm2,…, ρm,m−1), αi = vec(Ai), i = l,…,q, and βj = vec(Bj), j = 1,…, p. The parameter space is a subspace Θ of

c11ue051_fmt

The true parameter valued is denoted by

c11ue052_fmt

Before detailing the estimation procedure and its properties, we discuss the conditions that need to be Imposed on the matrices Ai and Bj in order to ensure the uniqueness of the parameterization.

11.4.1 Identifiability Conditions

Let c11-ie292001_fmt and c11-ie292002_fmt. By convention, F100_fmtθ(z) = 0 if q = 0 and 212C_fmtθ(z) = Im if p = 0.

If 212C_fmtθ(z) is nonsingular, that is, If the roots of det (212C_fmtθ (z)) = 0 are outside the unit disk, we deduce from c11-ie292003_fmt the representation

(11.42) c11e042_fmt

In the vector case, assuming that the polynomials c11-ie293001_fmt and c11-ie293002_fmt have no common root is insufficient to ensure that there exists no other pair (F100_fmtθ, 212C_fmtθ), with the same degrees (p, q), such that

(11.43) c11e043_fmt

This condition is equivalent to the existence of an operator U(B) such that

(11.44) c11e044_fmt

this common factor vanishing in 212C_fmtθ(B)−1 F100_fmtθ(B) (Exercise 11.2).

The polynomial U(B) is called unimodular if det{U(B)} is a nonzero constant. When the only common factors of the polynomials P(B) and Q(B) are unimodular, that is, when

c11ue053_fmt

then P(B) and Q(B) are called left coprime.

The following example shows that, in the vector case, assuming that c11-ie293001_fmt(B) and c11-ie293002_fmt(B) are left coprime is insufficient to ensure that (11.43) has no solution θ ≠ θ0 (in the univariate case this is sufficient because the condition 212C_fmtθ(0) = c11-ie293002_fmt(0) = 1 imposes U(B) = U(0) = 1).

Example 11.4 (Nonidentifiable bivariate model) For m = 2, let

c11ue054_fmt

with

c11ue055_fmt

and

c11ue056_fmt

The polynomial F100_fmt(B) = U(B)c11-ie293001_fmt(B) has the same degree q as c11-ie293002_fmt(B), and 212C_fmt(B) = U(B)c11-ie293001_fmt(B) is a polynomial of the same degree p as c11-ie293002_fmt (B). On the other hand, U(B) has a nonzero determinant which is independent of B, hence it is unimodular. Moreover, 212C_fmt(0) = c11-ie293002_fmt(0) = Im and F100_fmt(0) = c11-ie293001_fmt(0) = 0. It is thus possible to find θ such that 212C_fmt(B) = 212C_fmtθ(B), F100_fmt(B) = F100_fmtθ(B) and omega-underline_fmt = U(1)omega-underline_fmt0. The model is thus nonidentifiable, θ and θ0 corresponding to the same representation (11.42).

Identifiability can be ensured by several types of conditions; see Reinsel (1997, pp. 37–40), Hannan (1976) or Hannan and Deistler (1988, sec. 2.7). To obtain a mild condition define, for any column i of the matrix operators F100_fmtθ(B) and 212C_fmtθ(B), the maximal degrees qi (θ) and pi(θ), respectively. Suppose that maximal values are imposed for these orders, that is,

(11.45) c11e045_fmt

where qiq and pip are fixed integers. Denote by aqi (i) (bpi (i)) the column vector of the coefficients of Bqi (Bpi) in the ith column of c11-ie293001_fmt(B) (c11-ie293002_fmt(B)).

Example 11.5 (Illustration of the notation on an example) For

c11ue057_fmt

with a11a21a12a22b11b21b12b22 ≠ 0, we have

c11ue058_fmt

and

c11ue059_fmt

Proposition 11.1 (A simple identifiability condition) If the matrix

(11.46) c11e046_fmt

has full rank m, the parameters α0 and β0 are identified by the constraints (11.45) with qi = qi0) and pi = pi0) for any value of i.

Proof. From the proof of the theorem in Hannan (1969), U(B) satisfying (11.44) is a unimodular matrix of the form U(B) = U0 + U1B + … + UkBk. Since the term of highest degree (column by column) of c11-ie293004_fmt, the ith column of F100_fmtθ(B) = U(B)c11-ie293001_fmt(B) is a polynomial in B of degree less than qi if and only if Ujaqi (i) = 0, for j = 1,…, k. Similarly, we must have Ujbpi (i) = 0, for j = 1,…, k and i = 1,… m. It follows that UjM(c11-ie293001_fmt, c11-ie293002_fmt) = 0, which implies that Uj = 0 for j = 1,…, k thanks to condition (11.46). Consequently U(B) = U0 and, since, for all θ, 212C_fmtθ(0) = Im, we have U(B) = Im.       Box_fmt

Example 11.6 (Illustration of the identifiability condition) In Example 11.4,

c11ue060_fmt

is not a full-rank matrix. Hence, the identifiability condition of Proposition 11.1 is not satisfied. Indeed, the model is not identifiable.

A simpler, but more restrictive, condition is obtained by imposing the requirement that

c11ue061_fmt

has full rank m. This entails uniqueness under the constraint that the degrees of F100_fmtθ and 212C_fmtθ are less than q and p, respectively.

Example 11.7 (Another illustration of the identifiability condition) Turning again to Example 11.5 with α12β21 = α22β11 and, for instance, α21 = 0 and α22 ≠ 0, observe that the matrix

c11ue062_fmt

does not have full rank, but the matrix

c11ue063_fmt

does have full rank.

More restrictive forms, such as the echelon form, are sometimes required to ensure identifiability.

11.4.2 Asymptotic Properties of the QMLE of the CCC-GARCH model

Let (2208_fmt1,…,2208_fmtn) be an observation of length n of the unique nonanticipative and strictly stationary solution (2208_fmtt) of model (11.41). Conditionally on nonnegative initial values 2208_fmt0,…, 2208_fmt1−q, c11-ie294001_fmtc11-ie294002_fmt−p the Gaussian quasi-likelihood is written as

c11ue064_fmt

where the c11-ie294003_fmt are recursively defined, for t ≥ 1, by

c11ue065_fmt

A QMLE of θ is defined as any measurable solution c11-ie291004_fmt such that

(11.47) c11e047_fmt

where

c11ue066_fmt

Remark 11.6 (Choice of initial values) It will be shown later that, as in the univariate case, the initial values have no influence on the asymptotic properties of the estimator. These initial values can be fixed, for instance, so that

c11ue067_fmt

They can also be chosen as functions of θ, such as

c11ue068_fmt

or as random variable functions of the observations, such as

c11ue069_fmt

where the first r = max{p, q} observations are denoted by 2208_fmt1−r,…, 2208_fmt0.

Let γ (A0) denote the top Lyapunov coefficient of the sequence of matrices A0 = (A0t) defined as in (11.36), at θ = θ0. The following assumptions will be used to establish the strong consistency of the QMLE.

A1: θ0 2208_fmt Θ and Θ is compact.

A2: γ (A0) < 0 and, for all θ 2208_fmt Θ, det 212C_fmt(z) = 0 21D2_fmt |z| > 1.

A3: The components of ηt are independent and their squares have nondegenerate distributions.

A4: If p > 0, then c11-ie293001_fmt(z) and c11-ie293002_fmt(z) are left coprime and M1(c11-ie293001_fmt, c11-ie293002_fmt) has full rank m.

A5: R is a positive definite correlation matrix for all θ 2208_fmt Θ.

If the space Θ is constrained by (11.45), that is, if maximal orders are imposed for each component of c11-ie291002_fmt and h_underline_t_fmt in each equation, then assumption A4 can be replaced by the following more general condition:

A4′: If p > 0, then c11-ie293001_fmt(z) and c11-ie293002_fmt(z) are left coprime and M(c11-ie293001_fmt, c11-ie293002_fmt) has full rank m.

It will be useful to approximate the sequence (l_carate_fmtt(θ)) by an ergodic and stationary sequence. Assumption A2 implies that, for all θ 2208_fmt Θ, the roots of 212C_fmtθ(z) are outside the unit disk. Denote by (h_underline_t_fmt)t = {h_underline_t_fmt(θ)}t the strictly stationary, nonanticipative and ergodic solution of

(11.48) c11e048_fmt

Now, letting Dt = {diag(h_underline_t_fmt)}½ and Ht = DtRDt, we define

c11ue070_fmt

We are now in a position to state the following consistency theorem.

Theorem 11.7 (Strong consistency) Let (c11-ie294004_fmt) be a sequence of QMLEs satisfying (11.47). Then, under A1–A5 (or Al–A3, A4′; and A5),

c11ue071_fmt

To establish the asymptotic normality we require the following additional assumptions:

A6: θ0 2208_fmtI_ring_fmt, where I_ring_fmt is the interior of Θ.

A7: c11-ie295001_fmt.

Theorem 11.8 (Asymptotic normality) Under the assumptions of Theorem 11.7 and A6–A7, root-n_fmt(c11-ie294004_fmtn − θ0) converges in distribution to 004E_fmt(0, J−1IJ−1), where J is A positive definite matrix and I is A positive semi-definite matrix, defined by

c11ue072_fmt

11.4.3 Proof of the Consistency and the Asymptotic Normality of the QML

We shall use the multiplicative norm (see Exercises 11.5 and 11.6) defined by

(11.49) c11e049_fmt

where A is a d1 × d2 matrix, 2016_fmtx2016_fmt is the Euclidean norm of vector c11-ie296001_fmt and ρ(·) denotes the spectral radius. This norm satisfies, for any d2 × d1 matrix B,

(11.50) c11e050_fmt

(11.51) c11e051_fmt

Proof of Theorem 11.7

The proof is similar to that of Theorem 7.1 for the univariate case.

Rewrite (11.48) in matrix form as

(11.52) c11e052_fmt

where B is defined in Corollary 11.1 and

(11.53) c11e053_fmt

We shall establish the intermediate results (a), (c) and (d) which are stated as in the univariate case (see Section 7.4 of Chapter 7), result (b) being replaced by

(b)′ {h_underline_t_fmt(θ) = h_underline_t_fmt0) P_theta_0_fmt a.s. and R(θ) = R0)} 21D2_fmt θ = θ0.

Proof of (a): initial values are asymptotically irrelevant. In view of assumption A2 and Corollary 11.1, we have ρ(double-struck-B_fmt) < 1. By the compactness of Θ, we even have

(11.54) c11e054_fmt

Iteratively using equation (11.52), as in the univariate case, we deduce that almost surely

(11.55) c11e055_fmt

where h_underline_t_fmtt, denotes the vector obtained by replacing the variables c11-ie296002_fmt by c11-ie296003_fmt in Ht,. Observe that K is a random variable that depends on the past values {2208_fmtt,t ≤ 0}. Since K does not depend on n, it can be considered as A constant, such as ρ. From (11.55) we deduce that, almost surely,

(11.56) c11e056_fmt

Noting that 2016_fmtR−12016_fmt is the inverse of the eigenvalue of smallest modulus of R, and that c11-ie297001_fmt, we have

(11.57) c11e057_fmt

using A5, the compactness of Θ and the strict positivity of the components of c11-ie297002_fmt. Similarly, we have

(11.58) c11e058_fmt

Now

(11.59) c11e059_fmt

The first sum can be written as

c11ue073_fmt

as η → ∞, using (11.51), (11.56), (11.57), (11.58), the Cesàro lemma and the fact that ρt2016_fmt2208_fmtt2208_fmtt2016_fmt = ρt2208_fmtt2208_fmtt → 0 a.s.5 Now, using (11.50), the triangle inequality and, for x ≥ − 1, log(l + x) ≤ x, we have

c11ue074_fmt

and, by symmetry,

c11ue076_fmt

Using (11.56), (11.57) and (11.58) again, we deduce that the second sum in (11.59) tends to 0. We have thus shown that almost surely as n → ∞,

c11ue077_fmt

Proof of (b)′: identifiability of the parameter. Suppose that, for some θ ≠ θ0,

c11ue078_fmt

Then it readily follows that ρ = ρ0 and, using the invertibility of the polynomial 212C_fmtθ(B) under assumption A2, by (11.42),

c11ue079_fmt

that is,

c11ue080_fmt

Let c11-ie298001_fmt. Noting that F10F_fmt0 = F10F_fmt(0) = 0 and isolating the terms that are functions of ηt−1

c11ue081_fmt

where Zt−2 belongs to the σ -field generated by {ηt−2t−3 …}. Since ηt−1 is independent of this σ-field, Exercise 11.3 shows that the latter equality contradicts A3 unless, for i, j = 1, …, m, pijhjj,t = 0 almost surely, where the pij are the entries of F10F_fmt1. Because hjj,t − 0 for all j, we thus have F10F_fmt1 = 0. Similarly, we show that F10F_fmt(B) = 0 by successively considering the past values of ηt−1. Therefore, in view of A4 (or A4′), we have α = α0 and β = β0 (see Section 11.4.1). It readily follows that c11-ie298002_fmt. Hence θ = θ0 We have thus established (b)′.

Proof of (c): the limit criterion is minimized at the true value. As in the univariate case, we first show that E_theta-0_fmt2113_fmtt(θ) is well defined on 211D_fmt 22C3_fmt {+∞} for all θ, and on 211D_fmt forθ = θ0 We have

c11ue082_fmt

At θ0, Jensen’s inequality, the second inequality in (11.50) and Corollary 11.2 entail that

c11ue083_fmt

It follows that

c11ue084_fmt

Because c11-ie299001_fmt, the existence of c11-ie299002_fmt in 211D_fmt holds. It is thus not restrictive to study the minimum of c11-ie299003_fmt for the values of θ such that c11-ie299004_fmt. Denoting by λi,t, the positive eigenvalues of c11-ie299005_fmt (see Exercise 11.15), we have

c11ue085_fmt

because log xx − 1 for all x > 0. Since logx = x − 1 if and only if x = 1, the inequality is strict unless if, for all i, λit = 1 P_theta_0_fmt -a.s., that is, if Ht(θ) = Ht0), P_theta_0_fmt -a.s. (by Exercise 11.15). This equality is equivalent to

c11ue086_fmt

and thus to θ = θ0, from (by)′.

Proof of (d). The last part of the proof of the consistency uses the compactness of Θ and the ergodicity of (2113_fmtt(θ)), as in the univariate case.

Theorem 11.7 is thus established.       Box_fmt

Proof of Theorem 11.8

We start by stating a few elementary results on the differentiation of expressions involving matrices. If f(A) is a real-valued function of a matrix A whose entries aij are functions of some variable x, the chain rule for differentiation of compositions of functions states that

(11.60) c11e060_fmt

Moreover, for A invertible we have

(11.61) c11e061_fmt

(11.62) c11e062_fmt

(11.63) c11e063_fmt

(11.64) c11e064_fmt

(11.65) c11e065_fmt

(11.66) c11e066_fmt

(a) First derivative of the criterion. Applying (11.60) and (11.61), then (11.62), (11.63) and (11.64), we obtain

(11.67) c11e067_fmt

for i = 1,…, s1 = m + (p + q)m2, and using (11.65),

(11.68) c11e068_fmt

for i = s1 + 1,…, s0. Letting D0t = Dt0), R0 = R0),

c11ue087_fmt

and c11-ie300001_fmt, the score vector is written as

(11.69) c11e069_fmt

for i = 1,…, s1, and

(11.70) c11e070_fmt

for i = s1 + 1,…, s0.

(b) Existence of moments of any order for the score. In view of (11.51) and the Cauchy-Schwarz inequality, we obtain

c11ue088_fmt

for i, j = 1,…, s1,

c11ue089_fmt

for i = 1,…, s1 and j = s1 + 1,…, s0, and

c11ue090_fmt

for i, j = s1 + 1,…, s0. Note also that

c11ue091_fmt

To show that the score admits a second-order moment, it is thus sufficient to prove that

c11ue092_fmt

for all i1 = 1,…, m, all i = 1,…, s1 and r0 = 2. By (11.52) and (11.54),

c11ue093_fmt

and, setting s2 = m + qm2,

c11ue094_fmt

On the other hand, we have

c11ue095_fmt

where double-struck-B_fmt(i) = ∂double-struck-B_fmt/∂θi; is a matrix whose entries are all 0, apart from a 1 located at the same place as θi in double-struck-B_fmt. In an abuse of notation, we denote by Ht(i1) and h_underline_0t_fmt(i1) the i1th components of Ht and h_underline_t_fmt0). With arguments similar to those used in the univariate case, that is, the inequality x/(1 + x) ≤ xs for all x ≥ 0 and s 2208_fmt [0, 1], and the inequalities

c11ue096_fmt

and, setting c11-ie301001_fmt,

c11ue097_fmt

we obtain

c11ue098_fmt

where the constants ρj1 (which also depend on i1, s and r0) belong to the interval [0, 1). Noting that these inequalities are uniform on a neighborhood of θ0 2208_fmtI_ring_fmt, that they can be extended to higher-order derivatives, as in the univariate case, and that Corollary 11.2 implies that c11-ie301002_fmt, we can show a stronger result than the one stated: for all i1 = 1,…, m, all i, j, k = 1,…, s1 and all r0 ≥ 0, there exists a neighborhood F115_fmt0) of θ0 such that

(11.71) c11e071_fmt

(11.72) c11e072_fmt

and

(11.73) c11e073_fmt

(c) Asymptotic normality of the score vector. Clearly, {∂2113_fmtt0)/∂θ}t is stationary and ∂2113_fmtt0)/∂θ is measurable with respect to the σ-field 2131_fmtt generated by {ηu, u ≤ t}. From (11.69) and (11.70), we have E {∂2113_fmtt0)/∂θ|2131_fmtt−1} = 0. Property (b), and in particular (11.71), ensures the existence of the matrix

c11ue099_fmt

It follows that, for all c11-ie302001_fmt, the sequence c11-ie302002_fmt is an ergodic, stationary and square integrable martingale difference. Corollary A.1 entails that

c11ue100_fmt

(d) Higher-order derivatives of the criterion. Starting from (a) and applying (11.60) and (11.65) several times, as well as (11.66), we obtain

c11ue101_fmt

where

c11ue102_fmt

and c3 is obtained by permuting 2208_fmtt2208_fmtt and R−1 in C1. We also obtain

c11ue103_fmt

where

c11ue104_fmt

and c5 is obtained by permuting 2208_fmtt2208_fmtt and ∂Dt/∂θi in C4. Results (11.71) and (11.72) ensure the existence of the matrix J := E22113_fmtt0)/∂θ∂θ′, which is invertible, as shown in (e) below. Note that with our parameterization, ∂2R/∂θi∂θj = 0.

Continuing the differentiations, it can be seen that c11-ie303001_fmt is also the trace of a sum of products of matrices similar to the Ci. The integrable matrix 2208_fmtt2208_fmtt appears at most once in each of these products. The other terms are, on the one hand, the bounded matrices R−1, ∂R/∂θi and c11-ie303002_fmt and, on the other hand, the matrices c11-ie303003_fmt and c11-ie303004_fmt. From (11.71)–(11.73), the norms of the latter three matrices admit moments at any orders in the neighborhood of θ0. This shows that

c11ue105_fmt

(e) Invertibility of the matrix J. The expression for J obtained in (d), as a function of the partial derivatives of Dt and R, is not in a convenient form for showing its invertibility. We start by writing J as a function of Ht and of its derivatives. Starting from

c11ue106_fmt

the differentiation formulas (11.60), (11.63) and (11.65) give

c11ue107_fmt

and then, using (11.64) and (11.66),

c11ue108_fmt

From the relation Tr(A′B) = (vecA)′vecB, we deduce that

c11ue109_fmt

where, using vec(ABC) = (C2297_fmt A)vec B,

c11ue110_fmt

Introducing the m2 × S0 matrices

c11ue111_fmt

we have h = Hd with c11-ie304001_fmt Now suppose that J = Eh′h is singular. Then, there exists a nonzero vector c11-ie304002_fmt, such that c′Jc = Ec′h′hc = 0. Since c′h′hc ≥ 0 almost surely, we have

(11.74) c11e074_fmt

Because H2 is a positive definite matrix, with probability 1, this entails that dc = 0m2 with probability 1. Decompose c into c = (c1, c2)′ with c11-ie304003_fmt, where S3 = S0S1 = m(mm − l)/2. Rows 1, m + 1,…, m2 of the equations

(11.75) c11e075_fmt

give

(11.76) c11e076_fmt

Differentiating equation (11.48) yields

c11ue112_fmt

where

c11ue113_fmt

Because (11.76) is satisfied for all t, we have

c11ue114_fmt

where quantities evaluated at θ = θ0 are indexed by 0. This entails that

c11ue115_fmt

and finally, introducing a vector θ1 whose S1 first components are vec c11-ie304004_fmt,

c11ue116_fmt

by choosing c1 small enough so that θ1 2208_fmt Θ. If c1 ≠ 0 then θ1 ≠ θ0 This is in contradiction to the identifiability of the parameter, hence c1 = 0. Equations (11.75) thus become

c11ue117_fmt

Therefore,

c11ue118_fmt

Because the vectors ∂vecR/∂θi, i = s1 + 1,…, s0, are linearly independent, the vector c2 = c11-ie305001_fmt is nul, and thus c = 0. This contradicts (11.74), and shows that the assumption that J is singular is absurd.

(f) Asymptotic irrelevance of the initial values. First remark that (11.55) and the arguments used to show (11.57) and (11.58) entail that

(11.77) c11e077_fmt

and thus

(11.78) c11e078_fmt

From (11.52), we have

c11ue119_fmt

where r = max{p, q} and the tilde means that initial values are taken into account. Since c11-ie305002_fmt for all t > r, we have c11-ie305003_fmt and

c11ue120_fmt

Thus (11.54) entails that

(11.79) c11e079_fmt

Because

c11ue305001_fmt

we thus have (11.77), implying that

(11.80) c11e080_fmt

Denoting by h_underline_0t_fmt (i1) the i1th component of h_underline_t_fmt0),

c11ue121_fmt

where C0 is a strictly positive constant and, by the usual convention, the index 0 corresponding to quantities evaluated at θ = θ0. For a sufficiently small neighborhood F115_fmt0) of θ0, we have

c11ue122_fmt

for all i1, j1, j2 2208_fmt {1,…, m} and all δ > 0. Moreover, in h_underline_t_fmt(i1), the coefficient of c11-ie306001_fmt is bounded below by a constant c > 0 uniformly on θ 2208_fmt F115_fmt0). We thus have

c11ue123_fmt

for some ρ 2208_fmt [0, 1), all δ > 0 and all s 2208_fmt [0, 1]. Corollary 11.2 then implies that, for all r0 ≥ 0,

c11ue124_fmt

From this we deduce that

(11.81) c11e081_fmt

(11.82) c11e082_fmt

The last inequality follows from (11.77) because

c11ue125_fmt

By (11.67) and (11.68),

c11ue126_fmt

where

c11ue127_fmt

and C3 contains terms which can be handled as c1 and c2. Using (11.77)–(11.82), the Cauchy-Schwarz inequality, and

c11ue128_fmt

which follows from (11.71), we obtain

c11ue129_fmt

where ut is an integrable variable. From the Markov inequality, c11-ie306002_fmt, which implies that

c11ue130_fmt

We have in fact shown that this convergence is uniform on a neighborhood of θ0, but this is of no direct use for what follows. By exactly the same arguments,

c11ue131_fmt

where c11-ie307001_fmt is an integrable random variable, which entails that

c11ue132_fmt

It now suffices to observe that the analogs of steps (a)–(f) in Section 7.4 have been verified, and we are done.   Box_fmt

11.5 Bibliographical Notes

Multivariate ARCH models were first considered by Engle, Granger and Kraft (1984), in the guise of the diagonal model. This model was extended and studied by Bollerslev, Engle and Woolridge (1988). The reader may refer to Hafner and Preminger (2009a), Lanne and Saikkonen (2007), van der Weide (2002) and Vrontos, Dellaportas and Politis (2003) for the definition and study of FF-GARCH models of the form (11.26) where p is not assumed to be orthonormal. The CCC-GARCH model based on (11.17) was introduced by Bollerslev (1990) and extended to (11.18) by Jeantheau (1998). A sufficient condition for strict stationarity and the existence of fourth-order moments of the CCC-GARCH(p, q) is established by Aue et al. (2009). The DCC formulations based on (11.19) and (11.20) were proposed, respectively, by Tse and Tsui (2002), and Engle (2002a). The single-factor model (11.24), which can be viewed as a dynamic version of the capital asset pricing model of Sharpe (1964), was proposed by Engle, Ng and Rothschild (1990). The main references on the O-GARCH or PC-GARCH, models are Alexander (2002) and Ding and Engle (2001). See van der Weide (2002) and Boswijk and van der Weide (2006) for references on the GO-GARCH model. Hafner (2003) and He and Terasvirta (2004) studied the fourth-order moments of multivariate GARCH models. Dynamic conditional correlations models were introduced by Engle (2002a) and Tse and Tsui (2002). These references, and those given in the text, can be complemented by the recent surveys by Bauwens, Laurent and Rombouts (2006) and Silvennoinen and Terasvirta (2008), and by the book by Engle (2009).

Jeantheau (1998) gave general conditions for the strong consistency of the QMLE for multivariate GARCH models. Comte and Lieberman (2003) showed the consistency and asymptotic normality of the QMLE for the BEKK formulation. Asymptotic results were established by Ling and McAleer (2003a) for the CCC formulation of an ARMA-GARCH, and by Hafner and Preminger (2009a) for a factor GARCH model of the FF-GARCH form. Theorems 11.7 and 11.8 are concerned with the CCC formulation, and allow us to study a subclass of the models considered by Ling and McAleer (2003a), but do not cover the models studied by Comte and Lieberman (2003) or those studied by Hafner and Preminger (2009b). Theorems 11.7–11.8 are mainly of interest because that they do not require any moment on the observed process and do not use high-level assumptions. For additional information on identifiability, in particular on the echelon form, one may for instance refer to Hannan (1976), Hannan and Deistler (1988), Lültkepohl (1991) and Reinsel (1997).

Portmanteau tests on the normalized residuals of multivariate GARCH processes were proposed, in particular, by Tse (2002), Duchesne and Lalancette (2003).

Bardet and Wintenberger (2009) established the strong consistency and asymptotic normality of the QMLE for a general class of multidimensional causal processes.

Among models not studied in this book are the spline GARCH models in which the volatility is written as a product of a slowly varying deterministic component and a GARCH-type component. These models were introduced by Engle and Rangel (2008), and their multivariate generalization is due to Hafner and Linton (2010).

11.6 Exercises

11.1 (More or less parsimonious representations)

Compare the number of parameters of the various GARCH (p, q) representations, as a function of the dimension m.

11.2 (Identifiability of a matrix rational fraction)

Let c11-ie308001_fmt denote square matrices of polynomials. Show that

(11.83) c11e083_fmt

for all z such that det c11-ie308002_fmt if and only if there exists an operator U(z) such that

(11.84) c11e084_fmt

11.3 (Two independent nondegenerate random variables cannot be equal)

Let X and Y be two independent real random variables such that Y = X almost surely. We aim to prove that X and Y are almost surely constant.

1. Suppose that Var(X) exists. Compute Var(X) and show the stated result in this case.

2. Suppose that X is discrete and P(X = x1)P(X = X2) ≠ 0. Show that necessarily x1 = x2 and show the result in this case.

3. Prove the result in the general case.

11.4 (Duplication and elimination)

Consider the duplication matrix Dm and the elimination matrix c11-ie308003_fmt defined by

c11ue133_fmt

where A is any symmetric m × m matrix. Show that

c11ue134_fmt

11.5 (Norm and spectral radius)

Show that

c11ue135_fmt

11.6 (Elementary results on matrix norms)

Show the equalities and inequalities of (11.50)–(11.51).

11.7 (Scalar GARCH)

The scalar GARCH model has a volatility of the form

c11ue136_fmt

where the αi and βj are positive numbers. Give the positivity and second-order stationarity conditions.

11.8 (Condition for the Lp and almost sure convergence)

Let p 2208_fmt [1, ∞[ and let (un) be a sequence of real random variables of Lp such that

c11ue137_fmt

for some positive constant C, and some constant ρ in ]0, 1[. Prove that

c11ue138_fmt

to some random variable u of Lp.

11.9 (An average of correlation matrices is a correlation matrix)

Let R and Q be two correlation matrices of the same size and let p 2208_fmt [0, 1]. Show that pR + (1 − p)Q is a correlation matrix.

11.10 (Factors as linear combinations of individual returns)

Consider the factor model

c11ue139_fmt

where the βj are linearly independent. Show there exist vectors αj such that

c11ue140_fmt

where the c11-ie309001_fmt are conditional variances of the portfolios c11-ie309002_fmt. Compute the conditional covariance between these factors.

11.11 (BEKK representation of factor models)

Consider the factor model

c11ue141_fmt

where the βj are linearly independent, ωj > 0, aj ≥ 0 and 0 ≤ bj < 1 for j = 1, …, r. Show that a BEKK representation holds, of the form

c11ue142_fmt

11.12 (PCA of a covariance matrix)

Let X be a random vector of 211D_fmtm with variance matrix Σ.

1. Find the (or a) first principal component of X, that is a random variable C1 = u1X of maximal variance, where u1u1 = 1. Is C1 unique?

2. Find the second principal component, that is, a random variable C2 = u2X of maximal variance, where u2u2 = 1 and Cov(C1, C2) = 0.

3. Find all the principal components.

11.13 (BEKK-GARCH models with a diagonal representation)

Show that the matrices A(i) and B(j) defined in (11.21) are diagonal when the matrices Aik and Bjk are diagonal.

11.14 (Determinant of a block companion matrix)

If A and D are square matrices, with D invertible, we have

c11ue143_fmt

Use this property to show that matrix B in Corollary 11.1 satisfies

c11ue144_fmt

11.15 (Eigenvalues of a product of positive definite matrices)

Let A and B denote symmetric positive definite matrices of the same size. Show that AB is diagonalizable and that Its eigenvalues are positive.

11.16 (Positive definiteness of a sum of positive semi-definite matrices)

Consider two matrices of the same size, symmetric and positive semi-definite, of the form

c11ue145_fmt

where A11 and B11 are also square matrices of the same size. Show that if A22 and B11 are positive definite, then so is A + B.

11.17 (Positive definite matrix and almost surely positive definite matrix)

Let A by a symmetric random matrix such that for all real vectors c ≠ 0,

c11ue146_fmt

Show that this does not entail that A is almost surely positive definite.

1 The choice is then unique because to any positive definite matrix A, one can associate a unique positive definite matrix R such that A = R2 (see Harville, 1997, Theorem 21.9.1). We have R = P Λ½ P′, where Λ½ is a diagonal matrix, with diagonal elements the square roots of the eigenvalues of A, and P is the orthogonal matrix of the corresponding eigenvectors.

2 For two matrices A = (aij) and B = (bij) of the same dimension, A 2299_fmt B = (aijbij).

3 For instance, c11ue005_fmt More generally, for ij, the [(j − 1)m + i]th and [(i − 1)m + j]th rows of Dm equal the m(m + l)/2-dimensional row vector all of whose entries are null, with the exception of the [(j − 1)(mj/2) + i ]th, equal to 1.

4 If A = (aij) is an m × n matrix and B is an m′ × n′ matrix, A 2297_fmt B is the mm′ × nn′matrix admitting the block elements aijB.

5 The latter statement can be shown by using the Borel-Cantelli lemma, the Markov inequality and by applying Corollary 11.2: c11ue075_fmt

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
18.224.62.160