Scribd
Upload
29 views2 pages

Midterm 2019

This document outlines the topics and questions for a stochastic calculus midterm exam, including: 1) Calculating the probability a fund manager is skilled based on past returns. 2) Showing a stochastic process is Markov and calculating expectations and variances. 3) Deriving distributions from stochastic differential equations. 4) Properties of moment generating functions and how they transform for linear combinations of random variables.

Uploaded by

Uasdaf
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
29 views2 pages

Midterm 2019

This document outlines the topics and questions for a stochastic calculus midterm exam, including: 1) Calculating the probability a fund manager is skilled based on past returns. 2) Showing a stochastic process is Markov and calculating expectations and variances. 3) Deriving distributions from stochastic differential equations. 4) Properties of moment generating functions and how they transform for linear combinations of random variables.

Uploaded by

Uasdaf
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Stochastic Calculus

Midterm exam
Prof. D. Filipović, E. Hapnes
29.10.2019

Time: 9:15-12:00. Full credit is for the correct answer and explanation.
Exercise 1: A fund manager can be skilled (S = H) or unskilled (S = L). If the fund manager
is skilled, then the return of her fund is good, RG , with probability q > 12 and bad, RB , with
probability 1 − q. If the fund manager is unskilled, then the return is RG with probability
1 − q and RB with probability q. Your prior belief is P rob(S = H) = 12 . Define Pn (m) as the
probability that the fund manager is skilled if you have n observations where the return is RG
in m of them.

(a) Find an expression for Pn (m).

(b) Define the random variable M , which takes the values M = 0, 1, ..., n, as the number of
times you observe the return RG . For n = 1, show that E[P1 (M )|S = H] > 21 .

(c) Calculate E[P1 (M )].


q
Hint: Find your solutions in terms of ρ = (1−q)
.

10 points

Exercise 2: Let Xt and Yt for t = 1, 2, ... be sequences of independent and identically dis-
2
tributed random variables with means µX and µY , and variances σX and σY2 . Consider the
stochastic process
Zt = Y1 X1 + Y2 X2 + · · · + Yt−1 Xt−1 + Yt Xt .

(a) Show that Zt is a Markov process.

(b) Calculate E[Zt+1 |Zt ] and V ar(Zt+1 |Zt ).

10 points
Exercise 3: Let St have the dynamics

dSt = µSt dt + σSt dWt

where Wt is a Brownian motion, S0 = s > 0, and µ and σ > 0 are constants.

(a) Calculate d log(St ).

(b) Show that the distribution of log(St ) is normal. What is its mean and variance?

(c) Calculate E[1{St ≥K} ] for some constant K > 0. Here, 1{St ≥K} is an indicator function that
takes the value 1 when St ≥ K and 0 otherwise.

10 points

Exercise 4: Let X be a random variable with mean µ and moment generating function MX (t).

(a) Show that:

• MX (t) ≥ eµt .
• P (X ≥ α) ≤ e−αt MX (t) for 0 ≤ t and any α as long as the moment generating
function exists.

(b) Let Y = a + bX. Calculate the moment generating function of Y as a function of the
moment generating function of X and the parameters a and b.

(c) Let X ∼ N (0, 1) and Y = X 2 . Calculate the moment generating function of Y , and
evaluate E[Y ] and V ar(Y ).

10 points

You might also like