Chapter 1 Preliminaries
Idempotent: A matrix A is idempotent if AA=A
- The rank of an idempotent matrix is equal to its trace, i.e. rank(A)=trace(A)
Generalized Inverse: G is a generalized inverse of A if AGA=A
Quadratic form, positive definite, non-negative definite
- A is PD ⇒ A−1 exists
linear transformation
multivariate normal distribution
if y∼N(μ,Σ), then yd=Az+μ where z∼N(0,In×n) and A is an m×n matrix of rank n such that AA′=Σ
if y∼N(μ,In×n), then w=y′y∼χ2n(μ′μ/2)
Suppose Σ is an n×n positive definite matrix, A is an n×n symmetric matrix of rank m such that AΣ is idempotent (AΣAΣ=AΣ). Then y∼N(μ,Σ)⇒y′Ay∼χ2m(μ′Aμ/2)
If w∼χ2m(θ) then E(w)=m+2θ and Var(w)=2m+8θ
Non-central t distribution: y∼N(δ,1), w∼χ2m, y and w are independent then y√w/m has a non-central t distribution with m degrees of freedom and non-centrality parameter δ
Non-central F distribution: w1∼χ2m1(θ) and w2∼χ2m2. w1 and w2 are independent then w1/m1w2/m2 has a non-central F distribution with m1 numerator degrees of freedom, m2 denominator degrees of freedom and non-centrality parameter θ: w1/m1w2/m2∼Fm1,m2(θ).
- if u∼tm(δ), then u2∼F1,m(δ2/2).