I’ve gotten some recent questions regarding which lessons are likely to be of greater importance on the exam, so I’ve updated my thoughts, and here they are. You may find these especially useful if you are off to a very late start getting going with the S material. Ideally, we would like to be starting on sample exams within the next couple of weeks. Oh, and on that note, I do plan update the Sample Exam strategy video next.
I didn’t notice when the CAS updated this file:
http://www.casact.org/admissions/passmarks/examS.pdf
but the pass mark from the Fall 2016 sitting was 54/90 = 60%.
The pass mark from the Fall 2015 was 52.5/88, which is right at 60%. The pass mark for Spring 2016 was 55/90, 61%. So it seems clear that a pass mark right in that ballpark of 60% is what we are likely to see moving into the future.
So, let’s talk about how we should manage our preparation time for the Spring 2017 sitting. I’ve already covered section D, on GLM’s, in the video marked D.0.1 How to Handle GLM Spring 2017. I’ll talk about the rest of the seminar just down below.
Before I begin, I want to point out that the lessons in this seminar were designed to offer complete and detailed coverage of the syllabus topics and the source readings. As always, some of those topics will be more popular than others, and some parts of the source readings will never be tested at all.
Thoroughly learning everything that could be on the exam will give you the best possible chance of passing the exam – but it will also take a very long time. If you try to fully prepare for everything that could potentially appear but you run out of time, you might end up failing to master some of the topics that are most likely to appear – and that could lead to disaster.
What I’m trying to say is, take the following recommendations as more of a list of topics that you should be absolutely certain to master than a list of things that you can skip.
Use the following guidance regarding which sections to spend the most time on in the full knowledge that I do not write any of the exam questions for this exam, I do not know what will actually appear on the exam, and I am not guaranteeing that the following guesses will be perfectly accurate.
B.1 Poisson Processes (Syllabus A.1-A.3, 0 to 15%)
Past exams: 15/135, 11% of the total exam
B.1.1: 2 problems
B.1.2: 3 problems
B.1.3: 2 problems
B.1.4: 4 problems
B.1.5: 1 problem
B.1.6: 3 problems
It seems as though it is likely that we will regularly see exam questions from all of the lessons in section B.1, (and from C.3.2 slide 4). But this is a good thing, because most of these will be type I problems that you can work immediately in less than 6 minutes. Chances are one or maybe two of these will look a bit tricky at first read through, but even these will sort themselves out with a few minutes thought. Pay a bit of extra attention to the ones from the first three sittings of exam S that look a bit odd to you as compared to the seminar exercises.
B.2 Continuous/Multiple lives and decrements (Syllabus A.4-A.5, 4 to 16%)
Past exams: 8 problems/135, 6% of the total exam
B.2.1: 2 problems
B.2.2: 1 problem
B.2.3: 1 problem
B.2.5: 1 problem
B.2.6: 1 problem
B.2.7: 2 problems
This 6% total is still on the low side The Fall 2016 exam actually had 4 problems, which was 9% of that exam. So, it could well be that we will continue to see more like 4-5 problems showing up on the exam from B.2.
After the first sitting, the exam writers have started to branch out in the B.2 material. At this point, they have still missed B.2.4, B.2.8 and B.2.9, but I think it is just a matter of time until we see problems coming from these sections. For example, the Fall 2016 exam hit lessons B.2.3 and B.2.6 for the first time.
It may be worth noting that B.2.1-B.2.3 have always had one or two problems, as have B.2.5-B.2.9.
My guess is that we will typically have one or two questions from B.2.1-B.2.4, and that we will also usually see one or two questions from the rest of the syllabus content as covered in B.2.5-B.2.9. Focus most of your attention on B.2.1-B.2.4, but also plan to learn the ideas in B.2.5-B.2.9 if you have enough time, since we are likely to see one or two problems from these 5 lessons. I suspect that one or more of the problems from this section will fairly often use the shortcuts in lesson C.3.2.
B.3 Markov Chains (Syllabus A.6, 2-10%)
Past exams: 12 problems, 9% of the exam
B.3.1: 2 problems (Both Fall 2016)
B.3.2: 1 problem
B.3.3: 6 problem
B.3.4: 2 problems
B.3.9: 1 problem
So far, we have seen 4 problems on each of the first three exams, so it seems like we may be settling into a pattern there.
We have also seen 2 problems from B.3.3 on each exam so far. I don’t think there will always be two from this lesson, but it seems like a lesson that should receive extra attention. Certainly we will be very surprised if there is a sitting where nothing from B.3.3 comes up.
It looks like what we are seeing is 3 problems from B.3.1-B.3.3, and then one problem chosen from the B.3.4-B.3.9 collection. I’m surprised that B.3.5 hasn’t shown up yet, and I think it is also easy to write a good problem from B.3.7, so I fully expect to see those sections cropping up on an exam before too long.
I expect that the material on Markov Chain Monte Carlo in B.3.6 and the material on continuous probabilities in B.3.8 to be comparatively rare, since it difficult to write good exam questions from this material. If I were very pressed for time, I might watch the B.3.6 and B.3.8 lessons but not do anything else to prepare for the topics in those two lessons.
B.4 Life Contingent Cash Flows (Syllabus A.7, 2-8%)
Past exams: 5/135, 3.7%
B.4.1: 2 problems (both Fall 2016)
B.4.3: 1 problem
B.4.8: 2 problems (but really just the equivalence principle)
Suggestion: Watch all lessons, ignore all formulas except MII (in B.4.5), Focus on B.4.1, B.4.3 and maybe B.4.8’s basic notion of the equivalence principle. Keep in mind that there will probably not be more than two problems (maybe three if they try to increase the past low percentage from this area) on the exam for all of these lessons. You will almost certainly be able to get at least one of the problems just from knowing B.4.1 and B.4.3. Come back and work on some other areas here at the end after you’ve finished everything else.
Section C: Statistics (Syllabus section B.1 – B.4, 20-40%)
Past Exams: 47 problems 35%
This material has continued from the first sitting to be tested on the high side, I expect because it is more challenging to write enough good GLM problems to test that section on the high side of its percentage range.
The exam questions in this section have been fairly routine. If you watch the lessons in section C complete most of the exercises, and pay careful attention to the practice exams and the previous exam S sittings, you should do well on this section.
There have been 12 from C.1, 14 from C.2, 13 from C.3, and 8 from C.4. At this point nearly all of the lessons in C have seen questions, oddly skipping C.2.5 and C.2.6, and totally as expected skipping C.3.6.
The only thing I would consider watching without attempting to master is lesson C.3.6, on the Run test. This has never been tested, although it is covered in the assigned readings, and is only included in the seminar because there was an example problem released for exam ST on the topic. We want as many points from this section as possible, since the problems are very likely to look familiar, with few unexpected twists or content selections.
Section D, GLM (Syllabus sections C.1 – C.4, 25-40%)
Past exams: 37 Problems, 27%
There is enough to say here that I already made an entire video about it. See lesson D.0.1, How to handle GLM Spring 2016. It is worth noting that there was much more attention given to the D.3. material in the Fall 2016, and rather less low hanging fruit in the form of D.2 and AIC problems. I think we are still far from a clear view of where the exam writers want to be with this material. It’s not surprising that this material is getting another change in the upcoming transition to MAS-I next year. On this section, we want to do the best we can with the time that we have, without sinking our chances by shortchanging other sections where the problems are likely to be more routine.
Section E. Time Series (Syllabus section D.1-D.3, 5-10%)
Past Exams: 11 questions, 8% of the exam.
Eight of the eleven questions that have appeared came from lessons E.2.2- E.2.5. The other three came from E.1.3-E.1.4.
There was a change to this material for the Fall 2016, adding in regression, deterministic seasonals and bias correction in forecasting, the material in the E.3 lessons. I was surprised that no question appeared from this new material, and I do expect them to examine that material soon. If I were doing it, the correction factor would probably be my first choice, but also, categorical seasonals are a very nice overlap with the GLM content. I’m not sure there will be much emphasis on the R code aspects of the readings, since it is too easy to write good exam questions for this material without including those commands. My expectation is that we will see 3-4 problems from the material. I think we will nearly always see something from E.1.3-E.1.4, and something from E.2.2-E.2.5. The third and possibly fourth problem may wander around through the rest of the lessons.