

FlashChat  Actuarial Discussion  Preliminary Exams  CAS/SOA Exams  Cyberchat  Around the World  Suggestions 
DW Simpson 
Actuarial Salary Surveys 
Actuarial Meeting Schedule 
Contact DW Simpson 
Probability Old Exam P Forum 

Thread Tools  Search this Thread  Display Modes 
#1




Dave Revelle's video problem solving session
I will be teaching an online seminar for Exam 1/P with the Infinite Actuary beginning with the August sitting of the exam. In addition, I will post problems here most weeks along with links to video solutions. The topics of these problems will correspond to the recommended study schedule for the seminar once it gets underway, but before that happens the topics will be fairly arbitrarily.
To start, here are some problems on moment generating functions. I've chosen this topic partly because it is relatively late on most people's study schedules and the February sitting is getting close. These questions don't cover everything on the subject, so I will revisit mgf's some time in the future. If anyone has requests for other topics, let me know. ************************************************** 1. Suppose the moment generating function for X is given by for t<2. Find the variance of X. Link to solution at the end. Numerical answer: Spoiler: ************************************************** 2. Let , where is a random variable whose moment generating function is Find Link to solution at end. Answer: Spoiler: ************************************************ 3. Suppose X and Y are random variables whose joint density is given by for and otherwise. Find for and . Link to solution at end. Answer: Spoiler: ************************************************** ** Video solutions available at the Infinite Actuary Last edited by daaaave; 02122007 at 08:43 AM.. 
#2




Here are this week's problems. Video solutions are available at The Infinite Actuary
. 1. Suppose that and are random variables whose moment generating functions are given by Find the coefficient of variation of W=X+Y+Z. Answer: Spoiler: 2. A random variable has a discrete distribution with Find the moment generating function of Answer: Spoiler: 3. A bivariate random variable has joint moment generating function Find Cov(X_1,X_2). Answer: Spoiler: Detailed video solutions are now available at The Infinite Actuary. Last edited by daaaave; 02122007 at 08:43 AM.. 
#3




This week's problems are based on old practice material from before the 2005 syllabus change that I think is still relevant. As usual, numerical answers are here and video solutions will be available at The Infinite Actuary.
1. [S '01 #36] A town in the shape of a square with each side measuring 4 has an industrial plant at its center. The industrial plant is polluting the air such that the concentration of pollutants at each location in the town can be modeled by the function for and . Calculate the average pollution concentration over the entire town. Note: the average concentration is the same thing as asking about the expected value of C(X,Y) if (X,Y) are uniformly chosen over the town. Answer: Spoiler: 2. (#7 from 1999 practice exam for the old syllabus) As part of the underwriting process for insurance, each prospective policyholder is tested for high blood pressure. Let X represent the number of tests completed when the first person with high blood pressure is found. The expected value of X is 12.5. Calculate the probability that the sixth person tested is the first one with high blood pressure. Answer: Spoiler: 3. (Based on #10 from 1999 practice exam for the old syllabus, modified to avoid an ambiguity in the problem.) An insurance policy covers the two employees of ABC Company. The policy will reimburse ABC for no more than one loss per employee in a year. It reimburses the full amount of the loss up to an annual companywide maximum of 8000. The probability of an employee incurring a loss in a year is 40%. The probability that an employee incurs a loss is independent of the other employee’s losses. The amount of each loss is uniformly distributed on [1000, 5000]. Given that at least one of the employees has incurred a loss in excess of 2000, determine the probability that losses will exceed reimbursements. Answer: Spoiler: Video Solutions are available at The Infinite Actuary We've changed the URL from previous weeks and also added QuickTime versions of the videos. Last edited by daaaave; 02122007 at 08:42 AM.. 
#4




Here are this week's problems. The second problem is about covariance matrices because of the recent threads on the topic (see here and here), but while it is a legitimate topic, the fact is that some of the textbooks on the syllabus don't cover it and I think it is unlikely to come up. A variation of question #2 that is more likely to appear on the exam would be something in which you are told Var(X_i)=3 for all i, and Cov(X_i,X_j)=2 for all .
1. Let be a randomly chosen integer with . What is the probability that is not divisible by 7, 11, or 13? Answer: Spoiler: 2. The covariance matrix of is given by If for all , find Answer: Spoiler: 3. Let and be independent, continuous uniform random variables on the interval [1,5]. Find . Answer: Spoiler: Video solutions are available at The Infinite Actuary Last edited by daaaave; 02182007 at 03:22 PM.. 
#5




Since the February sitting of the exam was this past week, I am going to take a couple weeks off from posting practice problems so I can finish the practice exams that I am working on.

#6




Hey Daave,
On question 9c in Lesson 3 of Continuous distributions, couldn't that be taken as waiting exactly 45 minutes rather than at least 45 minutes. I nearly tried to do this as a Poisson until I thought about the section this was from. 
#7




In general, how do you know when using the shortcut that the second derivative of the ln(Mx(t)) at t=0 equals V(x) is advantageous to finding the first two moments and plugging into V(x) = E(x^2)  [E(x)]^2?

#8




If M(t) is of the from e^(something), then it will usually be easier to work with ln(M(t)). For example, if or then will be easier to work with.
It is also probably easier to use ln(M(t)) when you have M(t)=something^power, such as because ln(M(t)) will have a nice derivative to work with. The main time when you might not want to work with ln(M(t)) is when M(t) is a sum of pieces. Logs work better with products than sums, and if we have something like then M'(t) and M''(t) will be nice, but ln(M(t)) and its derivatives will be messy. 
#9




Thanks Dave. I guess I was more worried about when not to use it as opposed to when to use it. So thank you for including that in your response.

#10




Info on watching the videos on an iPhone or iPod Touch:
http://www.actuarialoutpost.com/actu...d.php?t=161231 
Tags 
the infinite actuary, tia 
Thread Tools  Search this Thread 
Display Modes  

