24.02.2013 Views

Optimality

Optimality

Optimality

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

138 D. M. Dabrowska<br />

2.3. Some auxiliary notation<br />

From now on we assume that the function EN(t) is continuous. We shall need some<br />

auxiliary notation. Define<br />

e[f](u, θ) = E{Y (u)[fα](Γθ(u), θ, Z)}<br />

E{Yi(u)α(Γθ(u), θ, Z)} ,<br />

where f(x, θ, Z), is a function of covariates. Likewise, for any two such functions, f1<br />

and f2, let cov[f1, f2](u, θ) = e[f1f T 2 ](u, θ)−(e[f1]e[f2] T )(u, θ) and var[f](u, θ) =<br />

cov[f, f](u, θ). We shall write<br />

e(u, θ) = e[ℓ ′ ](u, θ), ē(u, θ) = e[ ˙ ℓ](u, θ),<br />

v(u, θ) = var[ℓ ′ ](u, θ), ¯v(u, θ) = var[ ˙ ℓ](u, θ), ρ(u, θ) = cov[ ˙ ℓ, ℓ ′ ](u, θ),<br />

for short. Further, let<br />

(2.4)<br />

Kθ(t, t ′ ) =<br />

Bθ(t) =<br />

and define<br />

� �<br />

(2.5) κθ(τ) =<br />

� t∧t ′<br />

0 � t<br />

0<br />

0

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!