Scribd
Upload
40 views

M5A42 Applied Stochastic Processes Problem Sheet 6 Term 1 2010-2011

This document contains 3 problems about stochastic processes: 1. The first problem examines a diffusion process X(t) on the interval [0,1] with drift a and diffusion D. It asks to write the generator and Kolmogorov equations, solve for the transition density, show ergodicity, find the invariant distribution, and calculate the stationary autocorrelation. 2. The second problem considers a diffusion Xt on [0,L] with drift a(x) and diffusion b(x). It asks about the stationary Fokker-Planck equation with reflecting, periodic, and absorbing boundary conditions. 3. The third problem examines a multidimensional diffusion Xt on Rd with drift

Uploaded by

sajal27feb
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
40 views

M5A42 Applied Stochastic Processes Problem Sheet 6 Term 1 2010-2011

This document contains 3 problems about stochastic processes: 1. The first problem examines a diffusion process X(t) on the interval [0,1] with drift a and diffusion D. It asks to write the generator and Kolmogorov equations, solve for the transition density, show ergodicity, find the invariant distribution, and calculate the stationary autocorrelation. 2. The second problem considers a diffusion Xt on [0,L] with drift a(x) and diffusion b(x). It asks about the stationary Fokker-Planck equation with reflecting, periodic, and absorbing boundary conditions. 3. The third problem examines a multidimensional diffusion Xt on Rd with drift

Uploaded by

sajal27feb
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

M5A42 APPLIED STOCHASTIC PROCESSES PROBLEM SHEET 6 Term 1 2010-2011

1. Let a, D be positive constants and let X(t) be the diffusion process on [0, 1] with periodic boundary conditions and with drift and diffusion coefcients a(x) = a and b(x) = 2D, respectively. Assume that the process starts at x0 , X(0) = x0 . (a) Write down the generator of the process X(t) and the forward and backward Kolmogorov equations. (b) Solve the initial/boundary value problem for the forward Kolmogorov equation to calculate the transition probability density p(x, t|x0 , 0). (c) Show that the process is ergodic and calculate the invariant distribution ps (x). (d) Calculate the stationary autocorrelation function
1 1

E(X(t)X(0)) =
0 0

xx0 p(x, t|x0 , 0)ps (x0 ) dxdx0 .

2. Let Xt be a one dimensional diffusion process with drift and diffusion coefcients a(x) and b(x), 1 d respectively, posed on the interval [0, L], and let J(x, t) = a(x)p(x, t) 2 dx (b(x)p(x, t)) denote the probability current. (a) Assume that Xt is equipped with reecting boundary conditions. Show that the stationary Fokker-Planck equation can be written as J(x) = 0. Solve this equation to obtain the invariant distribution. (b) Calculate the invariant distribution when the boundary conditions are periodic. (c) Is the process Xt ergodic when the boundary conditions are absorbing?

3. Let Xt be a multidimensional diffusion process on Rd with drift vector a(x) = Ax where A Rdd is a symmetric positive denite matrix and diffusion matrix b(x) = 2DI, where D > 0 and I is the identity matrix. (a) Write down the generator, the stochastic differential equation and the Fokker-Planck equation for Xt . (b) Show that Xt is an ergodic Markov process and calculate the invariant distribution. (c) Solve the Fokker-Planck equation with initial conditions p(x, t|x0 , 0) = (x x0 ). (d) Calculate the stationary autocorrelation matrix E(Xt X0 ) = x x0 p(x, t|x0 , 0)ps (x0 ) dxdx0 .
Rd Rd

(e) Solve the stochastic differential equation for Xt .

You might also like