Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 000014

    Uploaded by

    wiertrass2016
    8 views2 pages
    The document contains instructions for a take-home final exam in stochastic processes due on December 18. It includes 4 questions related to Markov chains, Brownian motion, Brownian bridges, and diffusion processes. The questions involve deriving transition probabilities, determining limiting distributions, finding generator and semigroup representations, and identifying conditions for invariant probability measures.

    Original Description:

    Original Title

    [000014]

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    The document contains instructions for a take-home final exam in stochastic processes due on December 18. It includes 4 questions related to Markov chains, Brownian motion, Brownian bridges, and diffusion processes. The questions involve deriving transition probabilities, determining limiting distributions, finding generator and semigroup representations, and identifying conditions for invariant probability measures.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as pdf or txt
    8 views2 pages

    000014

    Uploaded by

    wiertrass2016
    The document contains instructions for a take-home final exam in stochastic processes due on December 18. It includes 4 questions related to Markov chains, Brownian motion, Brownian bridges, and diffusion processes. The questions involve deriving transition probabilities, determining limiting distributions, finding generator and semigroup representations, and identifying conditions for invariant probability measures.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as pdf or txt
    of 2

    Take Home Final.

    Due on Dec 18

    Q1. Consider a population of fixed size N that consists k types of size n1 (t), n2 (t), . . . , nk (t)
    repectively in generation t. At generation t + 1 the population reproduces according to
    the folowing rule. Each member of the new generation can be of type 1, 2 . . . , k with
    probabilities n1N(t) , n2N(t) , . . . , nkN(t) repectively. The types of the N members of generation
    t + 1 are chosen independently with these probabilities. This defines a Markov Chain with
    transition probabilities π(m, n) on the space EN consisting of n = (n1 , n2 , . . . , nk ) with
    n1 + n 2 + · · · + n k = N .
    a). Write down explicitly the transition probability π(m, n).
    b). For smooth functions f on Rk , show that the limit
    X n m
    (Af )(x) = lim N [f ( ) − f ( )] π(m, n)
    N →∞
    m →x
    n
    N N
    N

    exists and evaluate A explicitly.


    c). Consider the rescaled process x(t) = n(Nt)
    N
    starting at time 0 from aN = N n
    , with a
    distribution of PN,aN . Show that PN,aN is totally bounded and any limit P with aN → x is
    P to the martingale problem for A that lives on the simplex {x = (x1 , x2 , . . . , xk ) :
    a solution
    xi ≥ 0, xi = 1} and starts at time 0 from a ∈ S.
    d). Show that the solution to the martingale problem for A is unique, by verifying
    that the equation
    ∂f (t, x)
    = Af ; f (0, x) = f0 (x)
    ∂t
    has solution which is a polynomial in x = (x1 , x2 . . . , xk ) of degree n with coefficients that
    are smooth functions of t, provided the initial data f0 (x) is a polynomial of n.
    e). What happens to the process x(t) as t → ∞ under Px ? If one of many things can
    happen try to determine the respective probabilities as functions of x.
    f ). How will things change if between generations there is a possibility of mutation
    pi,j
    where each individual of type i can change its type
    P to j with a small probability N
    and remain the same type with prbability 1 − N1 j6=i pi,j ? (Different individuals act
    independently.)

    Q2. Let l(t) be the local time at the origin of the one dimensional Brownian motion β(t).
    Z t
    l(t) = δ(β(s))ds
    0

    Define
    τ (t) = {sup s : l(s) ≤ t}

    1
    Show that τ (t) is a right continuous process with independent increments and find its
    Levy-Khintchine representation.

    Q3. The Brownina bridge x(t) is the Gaussian process on [0, 1] with E[x(t)] = 0 and
    E[x(s)x(t)] = min(s, t) − st. Show that its distribution is the same as that of β(t) − tβ(1).
    d2
    Show that it is a Markov process, and in fact a diffusion process with generator 12 dx 2 +
    d
    b(t, x) dx . Determine b(t, x) explicitly. Show that the Brownian bridge is the conditional
    distribution of the Brownina motion on [0, 1] given that β(1) = 0. In other words the
    transition probability density q(s, x; t, y) of the Brownian bridge is given by

    p(s, x; t, y)p(t, y; 1, 0)
    q(s, x; t, y) =
    p(s, x; 1, 0)

    where
    1 (y − x)2
    p(s, x; t, y) = √ exp[− ]
    2πt 2(t − s)

    Q4. Under what conditions on the function h(r) will the diffusion process with generator

    1 x
    ∆ + h(r) < · ∇ >
    2 r
    p
    where (r = (x21 + · · · + x2d ), have an invariant probabiliy measure on Rd ? Assume that
    h(r) is a smooth bounded function of r with h(0) = 0.

    You might also like