Scribd
Upload
206 views2 pages

Quantitative Finance: 1 Probability and Statistics

This document contains exam questions on quantitative finance topics including: 1) Maximum likelihood estimation of parameters for the normal distribution based on a sample. 2) Properties of white noise processes and stationary ARMA processes, including computing covariances and variances. 3) The GJR-GARCH volatility model, including computing unconditional variances, and one, two, and j-step ahead forecasts of conditional variances.

Uploaded by

Vidaup40
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
206 views2 pages

Quantitative Finance: 1 Probability and Statistics

This document contains exam questions on quantitative finance topics including: 1) Maximum likelihood estimation of parameters for the normal distribution based on a sample. 2) Properties of white noise processes and stationary ARMA processes, including computing covariances and variances. 3) The GJR-GARCH volatility model, including computing unconditional variances, and one, two, and j-step ahead forecasts of conditional variances.

Uploaded by

Vidaup40
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

Quantitative Finance

Midterm
Oct 4, 2019

1 Probability and Statistics


1.1 Maximum Likelihood Estimator (3.0 points)
The Normal density function is
1 (x−µ)2
f (x) = √ e− 2σ 2
2πσ 2
where µ and σ 2 are the mean and variance. Suppose that you have a sample of T observation
X1 , X2 , ..., XT , and you want to use the method of maximum likelihood to estimate the parameters
µ and σ 2
1. (1.0 point)What is the log-likelihood function?
2. (1.0 point) What are the first-order conditions?(That is, what are the derivatives with respect
to µ and σ 2 ?
3. (1.0 point) What are the maximum likelihood estimates of µ and σ 2 ?

2 Stationary Processes
2.1 White Noise (3.0 points)
Any time series, Xt+1 can be decomposed into its conditional mean, E[Xt+1 ], and a remainder
process, εt+1

Xt+1 = E[Xt+1 ] + εt+1

1. (1.0 point) Show that εt+1 is a zero-mean white noise process


2. (1.0 point) Show that εt+1 has mean zero conditional on the information set available at time
t
3. (1.0 point) The white noise term εt+1 is uncorrelated with the conditional mean term, E[Xt+1 ]

2.2 ARMA process (7.0 points)


Consider the following stationary process:

Xt = φXt−1 + εt + θεt−4

where εt ∼ N (0, σ 2 )
1. (2.0 point) Show that the cov(Xt , εt−3 ) = φ3 σ 2 .[Hint: First, compute cov(Xt , εt−1 ) and use
this result to compute cov(Xt , εt−2 )]
2. (1.0 point) Compute γ0 = var(Xt )
3. (3.0 point) Prove that γj = φγj−1 + θφ4−j σ 2 for 1 ≤ j ≤ 4
4. (1.0 point) Show that γj = φγj−1 for j ≥ 5

1
3 Introduction to Volatility Modeling
3.1 GJR-GARCH model (7.0 points)
Consider the following model:

Rt = σt zt
2
σt+1 = ω + αRt2 + γIt Rt2 + βσt2

where It is an indicator variable (


1 Rt < 0
It =
0 Rt ≥ 0
iid
and ω > 0, α ≥ 0, β ≥ 0, γ ≥ 0 and zt ∼ N (0, 1)
2 2
1. (1.0 points) Show that Et [Rt+1 ] = σt+1
2. (1.0 points) Show that E[It zt2 ] = 1/2
3. (2.0 points) Given Rt is a stationary process, compute the unconditional variance of Rt [Hint:
σt2 is a stationary process]
4. (1.0 point) Find the one and two-step-ahead forecast of σt2 .
5. (2.0 point) Find the j-step-ahead forecast of σt2 .

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy