Math623_Lecture3_Spring2016

Math623_Lecture3_Spring2016 - Computational Finance Lecture...

Info icon This preview shows pages 1–9. Sign up to view the full content.

Computational Finance - Lecture 3 David Eliezer, lecture notes by Paul M. N. Feehan Lecturer in Mathematics Rutgers, The State University of New Jersey and Numerix, LLC Spring 2016 Rutgers University – Mathematics 623
Image of page 1

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

Outline Sources of bias in Monte-Carlo estimators
Image of page 2
Sources of bias in Monte-Carlo estimators As well as reducing variance in a Monte Carlo estimator, we shall also need to reduce its bias. Possible sources of bias include I Payoff discretization; I Model discretization; I Payoff non-linearity; I Pseudo-random number generator quality. Almost all payoffs are non-linear functions of S ( t ). Poor choices of pseudo-random number generators lead to poor simulations of the desired probability density functions.
Image of page 3

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

Bias due to payoff discretization I Payoff discretization is one source of bias in estimators used in Monte Carlo computation of prices of options with path-dependent payoffs. As a first example, discretization of continuous integrals may introduce bias. If we use Monte Carlo simulation to compute the price of an option depending on a continuous average , S Average ( T ) := 1 T Z T 0 S ( u ) du , there will be an error (and so a bias) introduced by any choice numerical integration method. For this reason, practitioners prefer to specify a discrete average when designing an option payoff, ˆ S Average ( T ) := m X i =1 S ( t i ) ,
Image of page 4
Bias due to payoff discretization II where 0 = t 0 < t 1 < · · · < t m = T and the prices, S ( t i ), are unambiguous (for example, daily closing prices). Alternatively, one can partially correct for this discrete monitoring bias using a correction due to Broadie, Glasserman, and Kou [ ? ]. As a second example, discrete monitoring of a continuous barrier may introduce bias. If we use Monte Carlo simulation to compute the price of a barrier option with continuous monitoring of a barrier B > K , E h e - rT ( S ( T ) - K ) + 1 { M ( T ) B } i , there will be a bias due to the fact that numerical schemes can can only simulate the discrete running maximum, ˆ M ( T ) := max 0 i m { S ( t i ) } .
Image of page 5

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

Bias due to payoff discretization III For this reason, practitioners again prefer to specify a discretely monitored barrier when designing an option payoff for a client. As an exercise, can you tell in which direction the estimator will be biased for each kind of barrier option? Alternatively, one can partially correct for this discrete monitoring bias using a correction due to Broadie, Glasserman, and Kou [ ? ]. See Glasserman [ ? , § 6.4] for an explanation of this idea in the case of payoffs depending on extremes (maxima or minima) or averages. In summary, the Terms and Conditions (T & C) sheet will specify discrete monitoring, but if it specifies continuous monitoring, we can either use a Broadie-Glasserman-Kou correction or offer to revise the T & C sheet.
Image of page 6
Bias due to payoff discretization IV As a third example, for a lookback call option, ( ¯ M ( T ) - K ) + , we would offer to replace the continuous running maximum, ¯ M ( T ) := max u [0 , T ] S ( u ) , by the discrete running maximum, ¯ M ( T ) := max i [0 , 1 ,..., N ] S ( t i ) .
Image of page 7

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

Model discretization error Model discretization error is introduced when we choose to
Image of page 8
Image of page 9
This is the end of the preview. Sign up to access the rest of the document.
  • Spring '16
  • Math, Normal Distribution, Randomness, Cumulative distribution function, Pseudorandom number generator, inverse transform method

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern