Actuarial Outpost Finding variance with a control variable
 Register Blogs Wiki FAQ Calendar Search Today's Posts Mark Forums Read
 FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions

 Salary Surveys Property & Casualty, Life, Health & Pension Health Actuary JobsInsurance & Consulting jobs for Students, Associates & Fellows Actuarial Recruitment Visit DW Simpson's website for more info. www.dwsimpson.com/about Casualty JobsProperty & Casualty jobs for Students, Associates & Fellows

 Investment / Financial Markets Old Exam MFE Forum

#1
09-16-2017, 01:53 AM
 alpast Non-Actuary Join Date: Aug 2017 College: UNLV Posts: 5
Finding variance with a control variable

A monte carlo valuation is run using the control variate method. The estimated variance of the variable being estimated is 8. The estimate of the variance of the control variable is 12. The estimated covariance is 9.
Determine the difference in the number of simulations required so that the estimated mean is within .05 of the true mean 90% of the time between the valuation of with the naive monte carlo methods and valuation using the control variate method.

So let X but the variable were estimating and Y the control. Im guessing that Var(X) = 8, Var(Y) = 12 and Cov(X,Y)=9.

my work: https://imgur.com/a/a6Gan

Why are they saying the control variance is 2 and not 2/n which is what the formula in the manual says it should be?
#2
01-11-2018, 10:58 PM
 BartimaeusOfUruk SOA Join Date: Nov 2017 Location: Des Moines, Iowa Studying for FAP Posts: 8 Blog Entries: 3

By definition:
y* = control variate
y-bar= Monte Carlo estimate for option y
x-bar= Monte Carlo estimate for option x
x = true price of option x

"The estimated variance of the variable being estimated is 8. The estimate of the variance of the control variable is 12. The estimated covariance is 9."
Means:
Var[x-bar] = 8
Var[y-bar] = 12
Cov[x-bar,y-bar] = 9

Not: Var(X) = 8, Var(Y) = 12 and Cov(X,Y)=9

Thus you don't divide by n.

 Tags monte carlo