FINM345A09Lecture5Corrected

# FINM345A09Lecture5Corrected - FinM 345/Stat 390 Stochastic...

This preview shows pages 1–6. Sign up to view the full content.

FinM 345/Stat 390 Stochastic Calculus, Autumn 2009 Floyd B. Hanson , Visiting Professor Email: [email protected] Master of Science in Financial Mathematics Program University of Chicago Lecture 5 (from Chicago) More Jump-Diffusion Calculus, Simple and Compound 6:30-9:30 pm, 26 October 2009, Kent 120 in Chicago 7:30-10:30 pm, 26 October 2009 at UBS Stamford 7:30-10:30 am, 27 October 2009 at Spring in Singapore Copyright c ± 2009 by the Society for Industrial and Applied Mathematics, and Floyd B. Hanson. FINM 345/Stat 390 Stochastic Calculus Lecture5–page1 Floyd B. Hanson

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

View Full Document
FinM 345 Stochastic Calculus: 5. More Jump-Diffusion Calculus — Simple & Compound: 5.1. Jump-Diffusion Rules and SDEs Continued: (ﬁnishing Chapt. 4 of the textbook) 5.1.1. Solution Simulations for Linear Jump-Diffusion SDEs with Constant Coefﬁcients: Upon merging and modifying the simulation algorithms for small time increments in Figure 7 on L1-p32 and using the cumulative sum of normal RNG Wiener increment approximations, together with the cumulative sum of uniform RNG Poisson increment approximations with the binomial RNG with n=1 to model the “Bernoulli” zero-one jump law, we show simulations of the ﬁnancially relevant linear jump-diffusion process with constant parameters solution (4.39) on L4-51 in Figure 5.1. FINM 345/Stat 390 Stochastic Calculus Lecture5–page2 Floyd B. Hanson
The basic simulation is performed on the approximate exponent increment Δ Y i ' ( μ 0 - σ 2 0 / 2)Δ t + σ 0 Δ W i +ln(1+ ν 0 P i , (5.1) corresponding to SDE (5.1), where Δ t = 0 . 001 for this MATLAB-generated ﬁgure, Δ W i ' DW( i ) , Δ t = Dt , where DW= sqrt (Dt) * randn (1 ,N ) ; and Δ P = binornd (1 0 Dt, 1 ) , approximating the zero-one jump law through Bernoulli form of the binomial RNG when the ﬁrst parameter is n = 1 and the second is the Poisson parameter ΔΛ = λ 0 Δ t . { In place of calls by random states to randn (1 ) , use serial calls to normrnd (0 , sqrt (Dt) , 1 ) . } FINM 345/Stat 390 Stochastic Calculus Lecture5–page3 Floyd B. Hanson

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

View Full Document
The sample path of the state exponent, YS , starting from a zero initial condition YS(1)=0 rather than ln( x 0 ) , for i =1: N , is approximated by YS( i + 1)=YS( i )+( μ 0 - σ 2 0 / 2) * Dt+ σ 0 * DW( i ) + log (1+ ν 0 ) * DP ( i ) , with t ( i + 1) = i * Dt . The sample path of the desired state, XS , is approximated by X ( t ( i + 1)) ' XS( i + 1) = x 0 * exp(YS( i + 1)) . The mean trajectory, XM , is given by E[ X ( t ( i + 1))] ' XM( i + 1) = x 0 * exp(( μ 0 + λ 0 * ν 0 ) t ( i + 1)) . FINM 345/Stat 390 Stochastic Calculus Lecture5–page4 Floyd B. Hanson
The mean trajectory is also displayed in the ﬁgure along with the upper XT exponential standard deviation estimate E[ X ( t ( i +1))] * V ( i +1) ' XT( i =XM( i * V ( i and lower XB exponential standard deviation estimate E[ X ( t ( i +1))] /V ( i ' XB( i i /V ( i , where the factor V ( i +1)=exp ± p Var[ Y ( t ( i +1))] ² =exp ³ q ( σ 2 0 + λ 0 * log 2 (1+ ν 0 )) t ( i ´ is the exponential of the standard deviation of the exponent process Y ( t ) in discrete form.

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

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 04/19/2010 for the course FIN 390 taught by Professor Hansen during the Fall '09 term at University of Illinois, Urbana Champaign.

### Page1 / 81

FINM345A09Lecture5Corrected - FinM 345/Stat 390 Stochastic...

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online