09StochChpater3

09StochChpater3 - Chapter 3 Stochastic Integration and Ito...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Chapter 3 Stochastic Integration and Ito Calculus 3.1 Wiener integral The most basic integration theory is based on the Riemann integral of a real- valued function on an interval [0 ,T ]. It is easy to extend the integrand to be a vector-valued or a Banach space-valued function. The Riemann Stielt- jes integral is an extension such that the integrator is a function of bounded variation. The Lebesgue integral is to allow the integrator to be a measure on a more general measure space. In this section, we will consider the Wiener integral R T f ( t ) dB ( t ), where f ( t ) is a real-valued function, integrating over the Brownian motion. Suppose f is a step function on [0 ,T ], f ( t ) = c i t i ≤ t < t i +1 , i = , 1 , ··· ,n- 1. We write Δ B i = B ( t i +1 )- B ( t i ), and define I ( f ) = Z T f ( t ) dB ( t ) = n- 1 X i =0 c i Δ B i . (3.1.1) 83 84 CHAPTER 3. STOCHASTIC INTEGRATION AND ITO CALCULUS Example 3.1.1 . Let f ( t ) be a step function on [0 , 3] that takes values- 1 , 1 , 2 on the intervals [0 , 1), [1 , 2) and [2 , 3] respectively. Then I ( f ) = Z 3 f ( t ) dB ( t ) = (- 1)( B (1)- B (0)) + 1 · ( B (2)- B (1)) + 2 · ( B (3)- B (2)) = B (1) + ( B (2)- B (1)) + 2 · ( B (3)- B (1)) ∼ N (0 , 1) + N (0 , 1) + N (0 , 4) (independent sum) = N (0 , 6) . Lemma 3.1.1. Let f be a step function on [0,T]. Then I ( f ) ∈ L 2 (Ω) , it is a normal r.v. with mean and variance σ 2 = E (( I ( f )) 2 ) = Z T | f ( t ) | 2 dt . Proof . Recall that if X 1 , ··· ,X n are independent and X i ∼ N ( μ i ,σ 2 i ), then a 1 X 1 + ··· + a n X n ∼ N ( a 1 μ 1 + ··· + a n μ n , a 2 1 σ 2 1 + ··· + a 2 n σ 2 n ) . It follows that I ( f ) is a normal r.v. with mean 0. For the variance, we have E ( I ( f ) 2 ) = E ( X i,j c i c j Δ B i Δ B j ) = X i c 2 i E ( (Δ B i ) 2 ) = X i c 2 i ( t i +1- t i ) = Z T | f ( t ) | 2 dt . / Lemma 3.1.2. Suppose f ∈ L 2 [0 ,T ] , then there exist a sequence of step func- tion { f n } converges to f a.e., and { I ( f n ) } ∞ n =1 is a Cauchy sequence in L 2 (Ω) . Proof . The first statement is well known. The second statement follows from E ( ( I ( f n )- I ( f m )) 2 ) = Z T | f n ( t )- f m ( t ) | 2 dt . / 3.1. WIENER INTEGRAL 85 Definition 3.1.3. For f ∈ L 2 [0 ,T ] , we define I ( f ) = lim n →∞ I ( f n ) where { f n } is as in Lemma 3.1.2, and call I ( f ) the Wiener integral of f . Clearly I ( f ) ∈ L 2 (Ω) by the completeness of L 2 (Ω) and the definition is independent of the choice of the subsequence { f n } . Proposition 3.1.4. For f ∈ L 2 [0 ,T ] , I ( f ) ∈ L 2 (Ω) is a normal r.v. with mean and variance σ 2 = k f k 2 = R T | f ( t ) | 2 dt . Proof . We need to use the fact that if X n ∼ N ( μ n ,σ n ) and if X n → X in probability (or in L 2 (Ω)), then X ∼ N ( μ,σ ) with μ = lim n →∞ μ n and σ = lim n →∞ σ n / Corollary 3.1.5. If f,g ∈ L 2 [0 ,T ] , then E ( I ( f ) I ( g )) = R T f ( t ) g ( t ) dt ....
View Full Document

This note was uploaded on 05/23/2010 for the course MATH 987 taught by Professor Cheung,cecilia during the Spring '08 term at CUHK.

Page1 / 24

09StochChpater3 - Chapter 3 Stochastic Integration and Ito...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online