Вы находитесь на странице: 1из 37

ACTL2102 Final Exam PASS Session

ACTL2102 Final Exam PASS Session

October 30, 2014

1/37

ACTL2102 Final Exam PASS Session


Plan

1 Lectures 1-2: Stochastic Processes and Discrete Time Markov Chains


Stuff You Should Know
Example Questions
2 Lecture 3: Exponential Distribution and the Poisson Process
Stuff You Should Know
Example Questions
3 Lectures 4-5: Continuous Time Markov Chains
Stuff You Should Know
Example Questions
4 Lectures 6-8: Time Series Mathematics
Stuff You Should Know
Example Questions
5 Lectures 8-10: Model Selection, Checking, and Prediction
Stuff You Should Know
Example Questions
6 Lectures 11-12: Pseudo-Random Continuous Walks
Stuff You Should Know
Example Questions
2/37

ACTL2102 Final Exam PASS Session


Lectures 1-2: Stochastic Processes and Discrete Time Markov Chains
Stuff You Should Know

Stuff You Should Know


Definitions:
Independent Increments
Stationary Increments
Markov Process

Transition Matrices
State Classifications
Mean Time in Transient States
Periodicity
Limiting Probabilities and their Conditions
Time Reversibility
Gamblers Ruin
Branching Processes
3/37

ACTL2102 Final Exam PASS Session


Lectures 1-2: Stochastic Processes and Discrete Time Markov Chains
Example Questions

PASS 3-4 (modified)

Specify the classes of the following Markov chain, and determine


for each class whether they are transient or recurrent:
1 2

3
3 0 0
1 0 1 0
2

1 2
0 0 1 3
4
4
0 0 12 12
2 Find the probability that the process will return to state 4
eventually if the process starts at state 4.
3

4/37

Find the expected number of times state 4 is visited if the


process starts at state 4.

ACTL2102 Final Exam PASS Session


Lectures 1-2: Stochastic Processes and Discrete Time Markov Chains
Example Questions

Tutorial 2-2
An unfeasibly large organisation has N employees. Each employee
has one one of three possible job classifications and changes
classification (independently) according to a Markov chain with
transition probabilities.

0.7 0.2 0.1


0.2 0.6 0.2
0.1 0.4 0.5

5/37

What percentage of employees are in each classification?

Is this process time-reversible?

ACTL2102 Final Exam PASS Session


Lectures 1-2: Stochastic Processes and Discrete Time Markov Chains
Example Questions

PASS 4-4

For the following branching process, calculate the probability of


extinction when the number Y of offspring of each individual
follows the following distribution:
1

P(Y = 0) = 31 , P(Y = 1) = 12 , P(Y = 2) =

Y Bin(2, 0.6)

1
6

For both questions, assume the starting population X0 = 1

6/37

ACTL2102 Final Exam PASS Session


Lectures 1-2: Stochastic Processes and Discrete Time Markov Chains
Example Questions

PASS 4-6

Han Solo moves among n + 1 star systems that are arranged in a


circle. At each system it moves either to the next system in the
clockwise direction with probability p or the counterclockwise
direction with probability q = 1 p. Starting at a specified
system, call it system 0, find the probability that all star systems
have been visited before revisiting system 0.

7/37

ACTL2102 Final Exam PASS Session


Lecture 3: Exponential Distribution and the Poisson Process
Stuff You Should Know

Stuff You Should Know

Exponential Distribution: Properties and Results


Definition of Counting and Poisson Process
Interarrival and Waiting Times
Sum and Thinning of Poisson Process
Non-Homogenous and Compound Poisson

8/37

ACTL2102 Final Exam PASS Session


Lecture 3: Exponential Distribution and the Poisson Process
Example Questions

PASS 5-3
The average number of automobiles entering a mountain tunnel
per minute period is 1. Excessive number of cars entering the
tunnel during a brief period of time produces a hazardous
situation. Assuming the Poisson process:

9/37

Find the probability that the number of automobiles entering


the tunnel during a 1 minute period exceeds 2?

What is the probability that the number of automobiles


entering the tunnel will be less than 3 during a 3-minute
period?

What is the probability that no car enters the tunnel during


the first 4 minutes?

ACTL2102 Final Exam PASS Session


Lecture 3: Exponential Distribution and the Poisson Process
Example Questions

PASS 5-6
An insurance portfolio contains policies for three categories of
policyholder: A, B and C. The number of claims in a year, N, on
an individual policy follows a Poisson distribution with mean .
Individual claim sizes are assumed to be exponentially distributed
with mean 4 and are independent from claim to claim. The
distribution of , depending on the category of the policyholder, is:
Category
A
B
C

Value of
2
3
4

Proportion of policyholders
20%
60%
20%

Denote by S the total amount claimed by a policyholder in one


year.
10/37

ACTL2102 Final Exam PASS Session


Lecture 3: Exponential Distribution and the Poisson Process
Example Questions

PASS 5-6 continued

11/37

Prove that E (S) = E [E (S|)].

Show that E (S|) = 4 and Var (S|) = 32.

Calculate E (S).

Calculate Var (S).

ACTL2102 Final Exam PASS Session


Lecture 3: Exponential Distribution and the Poisson Process
Example Questions

Tutorial 3-6

Events occur according to a nonhomogenoous Poisson process


whose mean value function is given by
m(t) = t 2 + 2t,

t0

What is the probability that n events occur between time t = 4


and t = 5?

12/37

ACTL2102 Final Exam PASS Session


Lecture 3: Exponential Distribution and the Poisson Process
Example Questions

Tutorial 3-9
Insurer A has a combined home insurance and landlords insurance
portfolio. The total number of claims for this portfolio is modelled
using a Poisson process with expected claims 300 per year. The
proportion of landlords insurance claims was 1/5 of the overall
claims. Insurer A sells its home insurance portfolio to insurer B.
Insurer B specialises in home insurance and has no landlords
insurance policies. The expected number of claims for the old
portfolio of insurer B was 120.

13/37

Define the processes of the number of claims from insurer A


for both the home insurance portfolio and landlords insurance
portfolio before the takeover.

Define the processes of the number of claims of both insurer


A and insurer B before and after the takeover of the home
insurance by insurer B.

ACTL2102 Final Exam PASS Session


Lectures 4-5: Continuous Time Markov Chains
Stuff You Should Know

Stuff You Should Know

Definitions and First Principles


The Q (Generator) Matrix and the embedded Markov Chain
Kolmogorov Equations and the P(s,t) transition matrix
Limiting Probabilities
Time Reversibility
Birth-Death Processes (rates, structure, mean time,
Kolmogorov equations, balance equations)
HSD models, and computing probabilities using integrals
(Maybe) Calculation methods

14/37

ACTL2102 Final Exam PASS Session


Lectures 4-5: Continuous Time Markov Chains
Example Questions

PASS 6-1
A 24 hour convenience store has three cashiers, Ralph, Vincent,
Benjamin. Customers arrive according to a Poisson process with
rate per hour and join the queue if there are already 3 customers
served. Each cashier serves customers with service times
exponential with mean 1 per hour.

15/37

Define a Markov chain to model the number of customers in


the store and write down the corresponding generator matrix
(the instantaneous transition rate matrix).

Given that there is n (n > 3) customers in the store, determine


an expression for the conditional probability that the number
of customers will remain unchanged over the next 15 minutes.

ACTL2102 Final Exam PASS Session


Lectures 4-5: Continuous Time Markov Chains
Example Questions

PASS 6-5 (modified)


Consider the following probability transition rate matrix with state
1, 2, 3, 4:

0.7 0.4 0.3


0
0.1 1 0.3
0.6

Q=
0.1 1.9 2.2 0.2
1.7 0.3 0.5 2.5
1
2

16/37

Find the embedded probability matrix.


Find the probability that the fifteen transition will be into
state 3, given we started in state 1 and the thirteenth
transition was into state 2.
If possible, find the long-run proportion of time spent in each
state.
Is this process time-reversible?

ACTL2102 Final Exam PASS Session


Lectures 4-5: Continuous Time Markov Chains
Example Questions

PASS 7-1

Consider the birth and death process with i = for i = 0, 1, 2, ...


and j = for j = 1, 2, 3, .... Denote Ti as the time spent in state
i before moving to state i + 1. Find E[Ti ].

17/37

ACTL2102 Final Exam PASS Session


Lectures 4-5: Continuous Time Markov Chains
Example Questions

PASS 7-7
Consider a health (H), sickness (S) and death (D) model for an
individual aged x > 0 with the following rates:
The rate at which a healthy individual becomes sick is 0.001x
The rate at which a sick individual recovers is 0.002x
The rate at which a healthy individual dies is 0.001x
The rate at which a sick individual dies is 0.002x

18/37

Give an expression for the probability that a sick 65 year old


individual stays sick for at least 1 year and then becomes
healthy and remain so till age 67.

Find the probability that a sick 65 year old individual remains


sick until he dies.

ACTL2102 Final Exam PASS Session


Lectures 4-5: Continuous Time Markov Chains
Example Questions

Tutorial 5-1 (modified)

Consider two machines both with an exponential lifetime 1/.

19/37

There is a single repairdrone that can service machines at a


rate . Set up the Kolmogorov backward equations. Also, set
up the forward equations.

Consider instead the case where the inevitable automation of


all our jobs has not yet occurred and the repairdrone is
instead a repairhuman. This repairhuman repairs at a rate
(t), where t is the time since breakfast. Set up the
Kolmogorov backward and forward equations.

ACTL2102 Final Exam PASS Session


Lectures 6-8: Time Series Mathematics
Stuff You Should Know

Stuff You Should Know

Classical decomposition model


Removing deterministic seasonality and deterministic trends
Integrated Time Series
SARIMA models (and their subsets)
Causality and Invertibility
Calculating ACF:
Linear Filter Method
Yule-Walker Equations

Calculating PACF

20/37

ACTL2102 Final Exam PASS Session


Lectures 6-8: Time Series Mathematics
Example Questions

PASS 8-1

Find a filter of the form 1 + B + B 2 + B 3 that passes linear


trends without distortion and that eliminates arbitrary seasonal
components of period 2.

21/37

ACTL2102 Final Exam PASS Session


Lectures 6-8: Time Series Mathematics
Example Questions

PASS 10-6

Compute the ACF and the PACF for the following AR(2) process
Xt = 0.6Xt1 + 0.2Xt2 + Zt where Zt W .N(0, 2 ).

22/37

ACTL2102 Final Exam PASS Session


Lectures 6-8: Time Series Mathematics
Example Questions

Tutorial 8-3

Consider the AR(2) process {Xt } satisfying:


Xt Xt1 2 Xt2 = Zt
For what values of is this a causal process?

23/37

ACTL2102 Final Exam PASS Session


Lectures 6-8: Time Series Mathematics
Example Questions

UKCT6 10/07-10 (modified)


The time series Xt is assumed to be stationary and to follow an
ARMA(2,1) process defined by:
Xt = 1 +

24/37

1
1
8
Xt1 Xt2 + Zt Zt1
15
15
7

Determine the roots of the characteristic polynomial, and


explain how their values relate to the stationarity of the
process.

Find the ACF for lags 0, 1, and 2.

Determine the mean and variance of Xt .

ACTL2102 Final Exam PASS Session


Lectures 8-10: Model Selection, Checking, and Prediction
Stuff You Should Know

Stuff You Should Know


Model selection
Looking at pictures
AIC, BIC (AKA SBC), AICc

Model parameter estimation (finding and using SACF and


SPACF)
Residual Analysis: Portmanteau/Ljung-Box
Testing for non-stationarity
Looking at the diagram
Dickey-Fuller and Augmented Dickey-Fuller tests

Cointegrated Time Series


Markov Property
Forecasting
25/37

Box-Jenkins
Best Linear Predictor

ACTL2102 Final Exam PASS Session


Lectures 8-10: Model Selection, Checking, and Prediction
Example Questions

Mock Exam Q6 & Q7

No. I cant be bothered typing them out. Just... open the .pdf for
yourself.

26/37

ACTL2102 Final Exam PASS Session


Lectures 8-10: Model Selection, Checking, and Prediction
Example Questions

PASS 12-5

Consider the following 2 time series:


Xt = 2t 2 + 3 + Zt
Yt = 3(t 1)2 + Zt

27/37

We say that (Xt , Yt ) are integrated of order d. Find d.

Are Xt and Yt cointegrated? If so, give the cointegration


vector. If not, give reasons why not.

ACTL2102 Final Exam PASS Session


Lectures 8-10: Model Selection, Checking, and Prediction
Example Questions

Tutorial 10-2 (Modified)

Explain briefly whether the following processes are Markov:

28/37

AR(4)

ARMA (1,1)

ACTL2102 Final Exam PASS Session


Lectures 8-10: Model Selection, Checking, and Prediction
Example Questions

Tutorial 10-3 (Modified)

Suppose {Xt } is a stationary time series with mean and ACF


(.). Find the best linear predictor of Xn+h of the form aXn + b.

29/37

ACTL2102 Final Exam PASS Session


Lectures 8-10: Model Selection, Checking, and Prediction
Example Questions

Tutorial 10-4 (Modified)

Consider an ARIMA(1,1,1) model.

30/37

Use the Box-Jenkins apporach to derive the one-step and


two-step ahead forecasts, assuming the parameter values are
known.

Evaluate the prediction variance Var( Xn+1 xn (1)), assuming


the parameter values are known.

ACTL2102 Final Exam PASS Session


Lectures 11-12: Pseudo-Random Continuous Walks
Stuff You Should Know

Stuff You Should Know


Definitions of Brownian Motion
Properties of Brownian Motion
Stochastic Differential Equations (!)
Stochastic Integrals (!)
Monte Carlo Simulation
Linear Congruential Formula
Inverse-Transform (Discrete and Continuous)
Accept-Reject (Discrete and Continuous)
Variance Reduction Techniques
Importance Sampling
Number of Simulations
31/37

ACTL2102 Final Exam PASS Session


Lectures 11-12: Pseudo-Random Continuous Walks
Example Questions

Lecture 11-30

Evaulate E[Bt8 ], where Bt is standard Brownian Motion.

32/37

ACTL2102 Final Exam PASS Session


Lectures 11-12: Pseudo-Random Continuous Walks
Example Questions

Bonus Question

Consider the geometric Brownian Motion


Y (t) = exp(t + Xt )
where Xt has the stochastic differential equation
dXt = dt + dBt
and where Bt is standard Brownian motion.
Express dYt as a stochastic differential equation.

33/37

ACTL2102 Final Exam PASS Session


Lectures 11-12: Pseudo-Random Continuous Walks
Example Questions

Another Bonus Question

34/37

Consider the linear congruential formula. Where a = 11,


c = 37, m = 100 and x0 = 85, generate 3 random numbers
on U[0, 1].

Use these numbers to generate samplings from a Bin(3, 0.5)


and an exponential distribution with mean 0.5.

If possible, use the exponential random variable samples


obtained to generate samplings from a Gamma(2, 3) random
variable. Use the Acceptance-Rejection method.

ACTL2102 Final Exam PASS Session


Lectures 11-12: Pseudo-Random Continuous Walks
Example Questions

PASS 13-5

Suppose you are simulating a set or normally distributed random


variables with mean 65 and standard deviation 15. Find the
number of simulations required so that you will be in a 0.5% band
of the true value, 95% of the time.

35/37

ACTL2102 Final Exam PASS Session


Lectures 11-12: Pseudo-Random Continuous Walks
Example Questions

Lazy Questions:

36/37

Explain how using antithetical variates reduces estimate


variance.

Explain how using control variates reduces estimate variance.

Explain how importance sampling can increase computational


efficiency for determining expectations.

ACTL2102 Final Exam PASS Session


Lectures 11-12: Pseudo-Random Continuous Walks
Example Questions

WERE DONE HERE!

GOOD LUCK!

37/37

Вам также может понравиться