Parameter estimation for stochastic equations with ... - samos-matisse
Parameter estimation for stochastic equations with ... - samos-matisse
Parameter estimation for stochastic equations with ... - samos-matisse
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
of (3.1). We get<br />
(5.2)<br />
Z t,s :=<br />
=<br />
∫ t ∫ s<br />
0 0<br />
∫ t ∫ s<br />
0<br />
0<br />
k α (t, v)k β (s, u) dX v,u<br />
k α (t, v)k β (s, u)b(X v,u ) dudv + M α,β<br />
t,s .<br />
5.3 Remark. Moreover, <strong>for</strong> α, β > 1 2<br />
, it follows from [6] and [16] that if we denote<br />
then it holds that<br />
(5.4) X t,s =<br />
Denote<br />
(5.5) R t,s = d<br />
dω α t<br />
Then we have the following:<br />
K α (t, v) = α(2α − 1)<br />
d<br />
dω β s<br />
∫ t ∫ s<br />
0<br />
0<br />
∫ t ∫ s<br />
0<br />
∫ t<br />
v<br />
r 2α−1 (r − v) α− 3 2 dr<br />
K α (t, v)K β (s, u) dZ u,v .<br />
0<br />
k α (t, v)k β (s, u)b(X v,u ) dudv.<br />
• For every (α, β) ∈ (0, 1) 2 and if b is Lipschitz, the sample paths of the process R<br />
given by (5.5) belong to L 2 ([0, 1] 2 , ω α ⊗ ω β ). This can be viewed in the same<br />
way as in Theorem 2.<br />
• Clearly, the process R is related to the process Q (4.9) by<br />
(5.6) R t,s = c α,β t α− 1 2 s<br />
β− 1 2 Qt,s .<br />
¿From (5.2) and (5.5) we obtain that<br />
(5.7) Z t,s = θ<br />
∫ t ∫ s<br />
0<br />
0<br />
R v,u dω β udω α v + M α,β<br />
t,s .<br />
and then the MLE <strong>for</strong> the parameter θ in (3.1) can be written as<br />
(5.8) θ t = −<br />
∫ t ∫ t<br />
0<br />
∫ t<br />
0<br />
0 R v,u dMu,v<br />
α,β<br />
∫ t<br />
0 R2 v,u dωudω β v<br />
α<br />
As a final remark, we derive an easier expression <strong>for</strong> the process R (or, equivalently,<br />
<strong>for</strong> the process Q) appearing in the expression of the MLE in the linear case.<br />
In this case we have<br />
(5.9) R t,s = d<br />
dω α t<br />
d<br />
dω β s<br />
∫ t ∫ s<br />
0<br />
0<br />
k α (t, v)k β (s, u)X v,u dudv.<br />
We need first a more suitable expression of the process R given by (5.9).<br />
.<br />
12