Search the whole station

# 离散状态随机过程代写 Discrete State随机过程代写

## Final Exam 526

### 1.(2pts+3pts+6pts+3pts+4pts)

This exercise is known in physics as the problem of Maxwell’s demon, which is a thought experiment that considers the possibility of a system that may violate the second law of thermodynamics. We imagine two urns in which there are two types of particles that are in constant motion. The first type of particle moves slowly, and the second moves quicky. In the right urn sits a demon who constructed a small door placed at the openning connecting the two urns. When the demon sees that a slow particle is about to escape the right urn, the demon shuts the door and prevents that particle from entering the left urn.

The illustration of our setup is in the figure. There are seven fast moving particles (in red) and six slow moving particles (in blue). In the image you can see the demon that just closed the door between the two urns blocking a slow particle from moving from the right to the left urn. In general, we will assume n particles total with q [0, 1] such that qn particles are fast and (1 q)n are slow. Eventually, all of the slow particles will be trapped in the right urn, a low entropy state (a high degree of order) and thus the system would violate the second law of thermodynamics.

Other than the interventions by the demon the dynamics proceed as follows, at each time step, we pick a particle uniformly at random and move it to the other urn. Of course when the demon intervenes (when we try to move a slow particle from the right urn to the left one), the usual dynamics are violated. We let Xk denote the number of fast particles in the left urn at time step k. The next 5 exercises are in increasing level of difficulty.

#### (1) 离散状态随机过程代写

Design the sample space Ω, the state space S and explain why each Xk is a random variable. Why is X = {Xk}k∈N a stochastic process?

(2) Show that why X = {Xk}kN is a Markov chain on the state space S and calculate the (Markov) transition probabilities of X.

(3) Compute E(Xk) for X0 = z for a fixed 0 z qn. Complute limk→∞ E(Xk) as k → ∞ and discuss the limiting expression for q (0, 1) and also the endpoints q = {0, 1}.

(4) Compute the stationary distribution of X = {Xk}kN.

(5) Let Lk be the total number of particles at time k in the left urn. Construct the sample and state space (Ω and S) for (Xk , Lk). Justify that (X, L) = {(Xk , Lk)}k∈N is a Markov chain and compute its transition probabilities. Compute the stationary distribution of (X, L).

In an election A receives 200 votes while B receivies only 100. Assume that the probability of getting a vote is identical (50%) for A and B. What is the probability that A is always ahead throughout the count?

### 3.(3pts+3pts) 离散状态随机过程代写

Consider a barbershop which have two barbers, each with his own barber chair. Suppose that there is room for at most 5 customers, with 2 in service and 3 waiting. Assume that potential customers arrive according to a Poisson process at rate 6 per hour. Customers arrive when the shop is full will leave without receiving service. Assume that the duration of each haircut is an independent exponential random variable with a mean of 15 minutes. Customers are served in a first-come first-served manner by the first available barber.

(a) Let Xt be the number of customers at time t. Write the generator matrix Q and find a stationary distribution. You can provide formulas for the values of πi , it is not necessary to compute the numerical values.

(b) What is the long-run proportion of time there are two customers in service plus two customers waiting? What is the long-run proportion of time that barber A is busy (assume one named A and the other B)?

### 5.(4pts+4pts)

Let K, Q, and N be independent Poisson processes with rates µ, θ > λ > 0, respectively.

(1) Prove for every k N and t 0 we have P(Nt k) = E(1{Qtk}Lt) where Lt = (λ/θ)Qt e(λθ)t.

(2) Prove that P(Nt ≥ Kt) ≤ e(√µ−√λ)2t for all t ≥ 0.

### 6. (3pts+3pts+3pts+3pts) 离散状态随机过程代写

Consider the Bachalier model St = σWt where W is a Brownian motion. Let XT = (ST K)+ = max{ST K, 0} be the payoff of a European call option with strike K > 0. The price of the option is

### 8.(6pts) 离散状态随机过程代写

Consider a process Xt satisfying the following properties:

(1) Xt has independent increments.

(2) For 0 s t

Xt Xs N(µ(t s), σ2(t s)).

(3) Sample paths of Xt are continuous functions of t.

(4) X0 = 0.

The next:

### Related recommendations

• #### 模块和表示论代写 MATH 5735代写 数学作业代写 数学代写

437

MATH 5735 - Modules and Representation Theory Assignment 1 模块和表示论代写 1. (9 marks) Recall that an integral domain is a commutative ring (with unity) that has no zero divisors. (a) Pro...

View details
• #### 计算数论作业代写 Mth 440/540代写 数学作业代写 数学代写

178

Mth 440/540 Homework 计算数论作业代写 (1) Part I a.) Use Mathematica to determine the 100th prime, the 1000th prime, and the 10,000th prime. b.) Use Mathematica to determine the prime (1) ...

View details
• #### 数学期末考试练习代写 Math 132代写 数学考试助攻 数学代写

109

Math 132 Final exam practice 数学期末考试练习代写 Directions. This is not an assignment to be turned in. These questions are meant to provide practice for the final exam. Directions. This ...

View details
• #### 数学final代考 Math 328 & Math 601代写 Math代写

524

Final Examination Math 328 & Math 601 数学final代考 Directions: • Write your names and student number on the top right-hand corner of this page. • Open this booklet only when directed to...

View details
1