31
D. Normal Mixture Models and Elliptical Models 1. Normal Variance Mixtures 2. Normal Mean-Variance Mixtures 3. Spherical Distributions 4. Elliptical Distributions QRM 2010 74

1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

  • Upload
    trinhtu

  • View
    240

  • Download
    6

Embed Size (px)

Citation preview

Page 1: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

D. Normal Mixture Models and Elliptical Models

1. Normal Variance Mixtures

2. Normal Mean-Variance Mixtures

3. Spherical Distributions

4. Elliptical Distributions

QRM 2010 74

Page 2: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

D1. Multivariate Normal Mixture Distributions

Pros of Multivariate Normal Distribution

• inference is “well known” and estimation is “easy”.

• distribution is given by µ and Σ.

• linear combinations are normal (→ VaR and ES calcs easy).

• conditional distributions are normal.

• For (X1, X2)> ∼ N2(µ,Σ),

ρ(X1, X2) = 0 ⇐⇒ X1 and X2 are independent.

QRM 2010 75

Page 3: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Multivariate Normal Variance Mixtures

Cons of Multivariate Normal Distribution

• tails are thin, meaning that extreme values are scarce in the normal

model.

• joint extremes in the multivariate model are also too scarce.

• the distribution has a strong form of symmetry, called elliptical

symmetry.

How to repair the drawbacks of the multivariate normal model?

QRM 2010 76

Page 4: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Multivariate Normal Variance Mixtures

The random vector X has a (multivariate) normal variance mixture

distribution if

Xd= µ +

√WAZ, (1)

where

• Z ∼ Nk(0, Ik);

• W ≥ 0 is a scalar random variable which is independent of Z; and

• A ∈ Rd×k and µ ∈ Rd are a matrix and a vector of constants,

respectively.

Set Σ := AA>. Observe: X|W = w ∼ Nd(µ, wΣ).

QRM 2010 77

Page 5: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Multivariate Normal Variance Mixtures

Assumption: rank(A)= d ≤ k, so Σ is a positive definite matrix.

If E(W ) <∞ then easy calculations give

E(X) = µ and cov(X) = E(W )Σ.

We call µ the location vector or mean vector and we call Σ the

dispersion matrix.

The correlation matrices of X and AZ are identical:

corr(X) = corr(AZ).

Multivariate normal variance mixtures provide the most useful

examples of elliptical distributions.

QRM 2010 78

Page 6: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Properties of Multivariate Normal Variance Mixtures

1. Characteristic function of multivariate normal variance mixtures

φX(t) = E(exp{it>X}

)= E

(E(exp{it>X}|W

))= E

(exp{it>µ− 1

2W t>Σt}

).

Denote by H the d.f. of W . Define the Laplace-Stieltjes transform

of H

H(θ) := E(e−θW ) =

∫ ∞0

e−θudH(u).

Then

φX(t) = exp{it>µ}H(

1

2t>Σt

).

Based on this, we use the notation X ∼Md(µ,Σ, H).

QRM 2010 79

Page 7: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Properties of Multivariate Normal Variance Mixtures

2. Linear operations. For X ∼Md(µ,Σ, H) and Y = BX + b,

where B ∈ Rk×d and b ∈ Rk, we have

Y ∼Mk(Bµ + b, BΣB>, H).

As a special case, if a ∈ Rd,

a>X ∼M1(a>µ,a>Σa, H).

Proof:

φY(t) = E(eit>(BX+b)

)= eit

>b φX(B>t)

= eit>(b+Bµ) H

(1

2t>BΣB>t

).

QRM 2010 80

Page 8: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Properties of Multivariate Normal Variance Mixtures

3. Density. If P [W = 0] = 0 then as X|W = w ∼ Nd(µ, wΣ),

fX(x) =

∫ ∞0

fX|W (x|w)dH(w)

=

∫ ∞0

w−d/2

(2π)d/2|Σ|1/2exp

{−(x− µ)>Σ−1(x− µ)

2w

}dH(w).

The density depends on x only through (x− µ)>Σ−1(x− µ).

QRM 2010 81

Page 9: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Properties of Multivariate Normal Variance Mixtures

4. Independence.

If Σ is diagonal, then the components of X are uncorrelated.

But, in general, they are not independent,

e.g. for X ∼M2(µ, I2, H),

ρ(X1, X2) = 0 ; X1 and X2 are independent.

Indeed, X1 and X2 are independent iff W is a.s. constant.

i.e. when X = (X1, X2)> is multivariate normally distributed.

QRM 2010 82

Page 10: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Examples of Multivariate Normal Variance Mixtures

Two point mixture

W =

{k1 with probability p,

k2 with probability 1− pk1, k2 > 0, k1 6= k2.

Could be used to model two regimes - ordinary and stress.

Multivariate t

W has an inverse gamma distribution, W ∼ Ig(ν/2, ν/2).

Equivalently, νW ∼ χ

2ν.

This gives multivariate t with ν degrees of freedom.

Symmetric generalised hyperbolic

W has a GIG (generalised inverse Gaussian) distribution.

QRM 2010 83

Page 11: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

The Multivariate t Distribution

Density of multivariate t

f(x) = kΣ,ν,d

(1 +

(x− µ)′Σ−1(x− µ)

ν

)−(ν+d)2

where µ ∈ Rd, Σ ∈ Rd×d is a positive definite matrix, ν is the

degrees of freedom and kΣ,ν,d is a normalizing constant.

• E (X) = µ.

• As E(W ) = νν−2, we get cov (X) = ν

ν−2Σ. For finite

variances/correlations, ν > 2.

Notation: X ∼ td(ν,µ,Σ).

QRM 2010 84

Page 12: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Bivariate Normal and t

x�

y

-4 -2 0 2�

4�

-4-2

02

4

x�

y

-4 -2 0 2�

4�

-4-2

02

4-4�

-2� 0

2�

4�

X�

-4

-2

0

2�

4�

Y

00.

050.

10.

150.

20.

25Z

-4�-2�

0

2�

4�

X�

-4

-2

0

2�

4�

Y

00.

10.

20.

30.

4Z

Left plot is bivariate normal, right plot is bivariate t with ν = 3.

Mean is zero, all variances equal 1 and ρ = −0.7.

QRM 2010 85

Page 13: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Fitted Normal and t3 Distributions

• ••

••

••

••

••

•••

••

••

••

••

•••

• ••

•••

••

••

••••

••

•••

• ••

••

••

• ••

•• •

••

• •

••

••

•• ••

••

••

•••

•• ••

• •••••

••

•••

••

• ••

••

••

••

••

••• ••

••

••

••

••

••

••

•••

••

••

• ••

•• ••

• • ••

••

••

•• ••

••

• •••

••

•• •• •• •

••

•• • ••

• •••

• ••

• •

•••

••• •

•••

••

••••

• •

••

•• •••

•• •

••

••

••

•• ••

•••

••

•••

••••

••

••

••

••

••

••

•• •

••• •

•• •

•• ••

••• •••

••

••

•• •• ••

••

•• •

••

•• •

• •••

••

••••••

••

••

• •

•• ••

•••

•••

••

••

••

•••• •

•• •• •

• ••

••

••

••• •

• •

••

••

•••

••

••

•••

•••

• • ••

••

••

••

••• •

• •

••

•••

••

••

••

••

• •

•••

••

•••• •

••

••

••

•• ••• •

• ••••

••

• •• •••

••

••

••

•••

•• ••

••

• •••

• •• •

••

••

••

••

•••

• •••

• •••••

••

•••

••

••

••

••

••

• ••

••

••

•••

••

••

• •••

•• •

••

•••

••••

••

• •

••

••

•• •

• ••

•• ••

••

••

••

•• •

••

••

• •••

• •• ••

••••

••

• ••

• ••••

••

•••

••

••

••

• •••

•• ••• •

••

••

•••

••

••

••

••

••

•••

•••

••

••

• •••••

••

••

•• •

•••

••

••

••

••

••• ••

••• •

•••

• ••

••

••• •

••

•••

•• • •••

••

•• •

• •

••

• •

••

••

••

••• •

• ••

•••••

••

• ••• ••

•••

••

••

•••

••

• •• ••

••••

••

•••

•••

••

•••• •

••••• ••

••

••

•• • •• •

••

••

•• •

•• • •

••

•••

••

• •

• •

••

••

••

•• •

••

••

• ••••

••

• •••

•• •

•••

••

••

•• •• •

•••

•••

•••

••

•••

••

•••

•• •

•• •••••

••

•• ••

••

••• •••

• ••

•• •

•••

•••

••• •

• ••

••

• •

••

•••

••

•••

•••

•••

••

••

• •

•••

•• • ••

••

••

••

••

••

••

•••

••• •

• ••

•••

••

••

•••

• •

••

• •

••

•• • •

••

•• ••

••

••• ••

••

••

••

••

••

••

••

••

• •

••

•••

•••

•• •

••

•••

• •

•• ••

• •

••

••

•••

•• •

••

•• ••••••

•••

•• ••

••

•• •

•••• •• •

•••

••

•••• •

••

• •

•• •

••

• ••••

••

••

•••

• •••

• ••

•••

••

• • •••• •

•••

••

•••

••

• •• •• •

••

• •

• •

•••

••

••

• ••• •

•• •

••

••

••

•• •

••

••••• •

•• ••

••

••

•••

•• •

• •

••

••

••••

••

••

• • •••

••

• •••

••

•••

••

• •

••

••••

••

•••

•••

• •

••

••

•• •••

••

••

••

••

• •

••

• •• •

••

••

••

••

•• •••

•• •• •

• •

• •••

••

• •

• •

•••

••

• •

• ••

••

••

••

• ••

•••• •

••

•••

•• ••

••

••

••

••

••

• •

••

•••

• •••

•••

•••

••

••

• •

• •

••

••

• •

•••

• •••

••

• ••

••

••

••

••

•••

••

• ••

••

•••

••

••

••

••

•••

•• •

••

• •

••

••

•••

•••

••••

• •

•••

••

••

•• ••••

••

••

•• •

••

•• •

•••• •

••

••

••

••

••

••

••••

••••

••

••

••

••••

••

••

••

BMW

SIE

ME

NS

-0.15 -0.10 -0.05 0.0 0.05 0.10-0.1

5-0

.10

-0.0

50

.00

.05

0.1

0

••

••• •••• •••

• •• ••

••

•••

• ••

•••

••

• ••

• •

• ••

•••

•••

• •

•• •

•••

• •

••

••

••••

••

••

••

• ••

••• •

•• •••

•••

•••

••• ••••

••

••

•• •••

••

••

•••

••

• ••

• ••

• ••

••••

••

••

••

• •

••

••

•• •• •

•••••• •

••• •

•••• •

•••

••

•••

••

• •••

••

••

••

••

••

••

••

•••

••

••

• •

•••

•••

••

••

•••••

•••

•• ••

••

••

•• • ••

•• ••

•••

••

••

• •

•••• ••••

••

••

•• •

••• •

•••

•••

••

•••

• •

•• •••

••

•••

•• ••

•••

••

•••

••••

• •

•••

•••

• ••

•• •

• ••

••••

••

• •••

• •

• •

••

••

•• •

••

•••

••

••

• ••• ••• •

•••

•••

• ••

•••

•••

••

••

• •

• ••• •

••

•••

• •• ••

••••••

••

•• • •••

• •

••

•• ••

•••••

••

••

•••

•• • •• ••

•• •

• ••

••

•••

•••

• •••••

••

••

•• •••

•••• •

•••

• •

••

••

••• •

• •

••• •

••

••

••

••

• •••

••

• •

••• ••

••

••

••

••

••

• •

••• •

••

••••

•••

••

••

•••••• •

•••

••

• •

•• •• •

••

• • •• ••

••

••••

•• •

••

••

••

•••

••• •

••

•• ••••• •

• ••

•••

•••

••

••••

••

• ••

•• •

••

•• •

•• •• • •

••

••

•••

••• •

• •

••

• ••

••

••••

••

••

••• • •

••

••

••

••

• ••

•••

••

• ••

•••

••

•••

• •••

••••

• ••

••

••

•••

•••

• •• ••• •

••

••••

• •

•••

••

•••

••

••

•• •

••

••

• ••••

••

• ••••

••

••

•• •

••

•• •••

••

• •••

• ••

••

••

••

••

••

•••

••

••

• ••

• •

••• •

• •

•• •

••• ••

••

•••

•••

••

••

••

• ••

••••

••

••

•• ••••

••

••

••

• ••

•••

••

•••

•••

••

••

••

• •

••

••

••• ••

••

•••

•••

••••• •

•• ••

• •

••••• ••

••

••• •• •••

••

•••

••

•• •

••

•• •

•••

••

••

•••

••

••

•• •••

•• •••

••

• ••••

• • ••

• •••• •

••

••

••

•• ••••

••

••

••

•••••

••

•• •

•• ••• •

••

• •••

••••

••

••

•• •

•••

••

•••

•••• ••

••

••

• •••

••

•• •

•• • •• ••

• •• •• ••••

••

•••

••

•• •

• •

• ••

••

• •

• •

••

••

• •

••••

••

••

••

••

••

•••••

•• ••

••••

••

••• •

••

• •

• •

• • ••

• ••

•• ••

••

•• •

••• ••

••••

• ••

•••

••

••• •• ••

••

• ••

••••

••••

• •

• •••••

•••

•• ••

••

••• •

•• • ••

•••

••

•••• •••

•••

•••• •

•••••

••

•••

••

•••

• •

• ••

•••

•• •

••

••

•• •

•••

•••

••

••

• ••

•••• •

••

••

•• ••

• •••

••

••

••

•••

••

••••

••

•••

•• ••

••••

••

•• •• •••

•••

• •

• •

••

••

••••

••

••

•• ••••

••

••

••••

•• •

• ••

••

•• •• • •

• • ••

• •• •

•••

••

•••

•••

• •

• ••

•••

• •

•• •••

••••

• •••

• ••

• • •

•• •• ••

••

••

•• ••••

••

••••

•••

••••• • •

••• •

• • •

•• ••

••

• •

••

•• •

•••

••

••••

••

• •••

•••

•• ••••

••

••

••• •

••

••

•••

•••

••••

•••

•••

••

••

• •••

•••

• ••

••

••

• ••

•••

•••

••

••

••

••

••

• ••

••

••

• •••

•• ••

••

••

•••• •

••

••

••

•• •

•••

••

•• •

••

••

••

••

••••

••••••

••

••

BMW

SIE

ME

NS

-0.15 -0.10 -0.05 0.0 0.05 0.10-0.1

5-0

.10

-0.0

50

.00

.05

0.1

0

Simulated data (2000) from models fitted by maximum likelihood to

BMW-Siemens data. Left plot is fitted normal, right plot is fitted t3.

QRM 2010 86

Page 14: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Simulating Normal Variance Mixture Distributions

To simulate X ∼Md(µ,Σ, H).

1. Generate Z ∼ Nd(000,Σ), with Σ = AA>.

2. Generate W with df H (with Laplace-Stieltjes transform H),

independent of Z.

3. Set X = µ +√WAZ.

QRM 2010 87

Page 15: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Simulating Normal Variance Mixture Distributions

Example: t distribution

To simulate a vector X ∼ td(ν,µ,Σ).

1. Generate Z ∼ Nd(000,Σ), with Σ = AA>.

2. Generate V ∼ χ2ν and set W = ν

V .

3. Set X = µ +√WAZ.

QRM 2010 88

Page 16: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Symmetry in Normal Variance Mixture Distributions

Elliptical symmetry means 1-dimensional margins are symmetric.

Observation for stock returns: negative returns (losses) have heavier

tails than positive returns (gains).

Introduce asymmetry by mixing normal distributions with different

means as well as different variances.

This gives the class of multivariate normal mean-variance mixtures.

QRM 2010 89

Page 17: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

D2. Multivariate Normal Mean-Variance Mixtures

The random vector X has a (multivariate) normal mean-variance

mixture distribution if

Xd= m(W ) +

√WAZ, (2)

where

• Z ∼ Nk(0, Ik);

• W ≥ 0 is a scalar random variable which is independent of Z; and

• A ∈ Rd×k and µ ∈ Rd are a matrix and a vector of constants,

respectively.

• m : [0,∞)→ Rd is a measurable function.

QRM 2010 90

Page 18: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Normal Mean-Variance Mixtures

Normal mean-variance mixture distributions add asymmetry.

In general, they are no longer elliptical and corr(X) 6= corr(AZ).

Set Σ := AA>. Observe:

X|W = w ∼ Nd(m(w), wΣ).

A concrete specification of m(W ) is m(W ) = µ +Wγ.

Example: Let W have generalized inverse Gaussian distribution to

get X generalised hyperbolic.

γ = 0 places us back in the (elliptical) normal variance mixture

family.

QRM 2010 91

Page 19: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

D3. Spherical Distributions

Recall that a map U ∈ Rd×d is orthogonal if UU> = U>U = Id.

A random vector Y = (Y1, . . . , Yd)> has a spherical distribution if

for every orthogonal map U ∈ Rd×d

Yd= UY.

Use ‖·‖ to denote the Euclidean norm, i.e. for t ∈ Rd,

‖t‖ = (t21 + · · ·+ t2d)1/2.

QRM 2010 92

Page 20: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Spherical Distributions

THEOREM The following are equivalent.

1. Y is spherical.

2. There exists a function ψ of a scalar variable such that

φY(t) = E(eit>Y) = ψ(‖t‖2), ∀t ∈ Rd.

3. For every a ∈ Rd,

a>Yd= ‖a‖Y1.

We call ψ the characteristic generator of the spherical distribution.

Notation: Y ∼ Sd(ψ).

QRM 2010 93

Page 21: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Examples of Spherical Distributions

• X ∼ Nd(000, Id) is spherical. The characteristic function is

φX(t) = E(eit>X) = exp

(−1

2t>t

).

Then X ∼ Sd(ψ) with ψ(t) = exp(−1

2t).

• X ∼Md(000, Id, H) is spherical, i.e. Xd=√WZ.

The characteristic function is

φX(t) = H

(1

2t>t

).

Then X ∼ Sd(ψ) with ψ(t) = H(12t).

QRM 2010 94

Page 22: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

D4. Elliptical distributions

A random vector X = (X1, . . . , Xd)> is called elliptical if it is an

affine transform of a spherical random vector Y = (Y1, . . . , Yk)>, i.e.

Xd= µ +AY,

where Y ∼ Sk(ψ) and A ∈ Rd×k, µ ∈ Rd are a matrix and vector of

constants, respectively.

Set Σ = AA>.

Example: Multivariate normal variance mixture distributions

Xd= µ +

√WAZ.

QRM 2010 95

Page 23: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Properties of Elliptical Distributions

1. Characteristic function of elliptical distributions

The characteristic function is

φX(t) = E(eit>X) = E(eit

>(µ+AY)) = eit>µψ

(t>Σt

)Notation: X ∼ Ed(µ,Σ, ψ).

We call µ the location vector, Σ the dispersion matrix and ψ the

characteristic generator.

Remark: µ is unique but Σ and ψ are only unique up to a positive

constant, since for any c > 0,

X ∼ Ed(µ,Σ, ψ) ∼ Ed(µ, cΣ, ψ

( ·c

))QRM 2010 96

Page 24: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Properties of Elliptical Distributions

2. Linear operations. For X ∼ Ed(µ,Σ, ψ) and Y = BX + b,

where B ∈ Rk×d and b ∈ Rk, we have

Y ∼ Ek(Bµ + b, BΣB>, ψ).

As a special case, if a ∈ Rd,

a>X ∼ E1(a>µ,a>Σa, ψ).

Proof:

φY(t) = E(eit>(BX+b)

)= eit

>b φX(B>t)

= eit>(b+Bµ)ψ

(t>BΣB>t

).

QRM 2010 97

Page 25: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Properties of Elliptical Distributions

3. Marginal distributions. For X ∼ Ed(µ,Σ, ψ), set

X1 = (X1, . . . , Xk)> and X2 = (Xk+1, . . . , Xd)

>

µ =

(µ1

µ2

)and Σ =

(Σ11 Σ12

Σ21 Σ22

).

Then

X1 ∼ Ek(µ1,Σ11, ψ) X2 ∼ Ed−k(µ2,Σ22, ψ).

QRM 2010 98

Page 26: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Properties of Elliptical Distributions

4. Conditional distributions. The conditional distribution of

X2|X1 = x1 is elliptical, but in general with a different

characteristic generator ψ.

In the special case of multivariate normality, the characteristic

generator remains the same.

QRM 2010 99

Page 27: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Properties of Elliptical Distributions

5. Convolutions. Let X and Y be independent and

X ∼ Ed(µ,Σ, ψ) Y ∼ Ed(µ, Σ, ψ).

If Σ = Σ then

X + Y ∼ Ed(µ + µ,Σ, ψ),

where ψ(u) := ψ(u)ψ(u).

QRM 2010 100

Page 28: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Properties of Elliptical Distributions

• The density of an elliptical distribution is constant on ellipsoids.

• Many of the nice properties of the multivariate normal are preserved.

In particular, all linear combinations a1X1 + . . .+ adXd are of the

same type.

• All marginal distributions are of the same type.

Two rvs X and Y (or their distributions) are of the same type if

there exist constants a > 0 and b ∈ R such that Xd= aY + b.

QRM 2010 101

Page 29: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

References

• [Barndorff-Nielsen and Shephard, 1998] (generalized hyperbolic

distributions)

• [Barndorff-Nielsen, 1997] (NIG distribution)

• [Eberlein and Keller, 1995] ) (hyperbolic distributions)

• [Prause, 1999] (GH distributions - PhD thesis)

• [Fang et al., 1990] (elliptical distributions)

• [Embrechts et al., 2002] (elliptical distributions in RM)

QRM 2010 102

Page 30: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

Bibliography

[Barndorff-Nielsen, 1997] Barndorff-Nielsen, O. (1997). Normal

inverse Gaussian distributions and stochastic volatility modelling.

Scand. J. Statist., 24:1–13.

[Barndorff-Nielsen and Shephard, 1998] Barndorff-Nielsen, O. and

Shephard, N. (1998). Aggregation and model construction for

volatility models. Preprint, Center for Analytical Finance, University

of Aarhus.

[Eberlein and Keller, 1995] Eberlein, E. and Keller, U. (1995).

Hyperbolic distributions in finance. Bernoulli, 1:281–299.

[Embrechts et al., 2002] Embrechts, P., McNeil, A., and Straumann,

D. (2002). Correlation and dependency in risk management:

QRM 2010 103

Page 31: 1.Normal Variance Mixtures 2.Normal Mean-Variance … · 2.Normal Mean-Variance Mixtures 3.Spherical Distributions 4.Elliptical Distributions QRM 2010 74. D1. ... expfit>Xg = E E

properties and pitfalls. In Dempster, M., editor, Risk Management:

Value at Risk and Beyond, pages 176–223. Cambridge University

Press, Cambridge.

[Fang et al., 1990] Fang, K.-T., Kotz, S., and Ng, K.-W. (1990).

Symmetric Multivariate and Related Distributions. Chapman &

Hall, London.

[Prause, 1999] Prause, K. (1999). The generalized hyperbolic model:

estimation, financial derivatives and risk measures. PhD thesis,

Institut fur Mathematische Statistik, Albert-Ludwigs-Universitat

Freiburg.

QRM 2010 104