For those of us who already know we’re doing this.
5 Likes
welp, it looks like I will unfortunately be joining you
1 Like
I will say that I thought I needed motivation to finally do this, and it needed to be more than “exam raise, step closer to FCAS, maybe a better job.” It needed to be something that wasn’t just personal, it was intensely personal.
I think I finally have that now.
1 Like
I decided that I should do this some time ago and now that I’ve just started, I can’t remember what I was thinking.
2 Likes
You were thinking “I feel pretty good about myself, maybe a little too good - how can I kneecap my self-esteem? I know, I’ll study for an actuarial exam!”
2 Likes
Just got back to back 5s and am not sure what to do next to improve. Any ideas aside from studying harder?
Luck of the draw for exam questions?
Think about the material. What in there would you say you know what, I’m really not that comfortable with this areas? Focus on those things, hammer it until you’re beyond comfortable with them. If there were questions last time that you say I kind of got it, but _____ then that’s something I’d spend time on.
One of my long-standing rants:
This is where not being able to discuss questions utterly pisses me off. I had 2 specific questions that I’m still *I have **no idea** what they wanted candidates to do with it* that, since we can’t discuss it. And, since the CAS is completely silent about it or even providing sample questions on the exam that might help as a guide, exams continue to be a giant game of Calvinball and everyone is just supposed to accept it’s fine and normal.
If speed getting through the exam has been a problem, focus on getting through questions more efficiently. I don’t say “figure out how to do them faster” because that doesn’t imply correctly; you want correct, and the easier way to get there is to be efficient.
1 Like
I went looking for changes, because Rising Fellow makes mention of something, and … well, …
Another CAS-related rant:
The search on the CAS website sucks ass. Like, truly sucks ass. It’s as if AI used AI to figure out the results, and then someone said “no, this is still too helpful - let’s throw out stuff that’s even moderately useful and leave the rest, then throw that into a hat and randomly draw out results.” Paying a pre-K kid $5 to organize results would be more effective.
Also, CAS information is getting to be even more terrible as usual. There’s changes to the exam outlines for October 2026. Cool, can you give everyone something that specifically calls it out? No, we’re just going to point you to the exam pages; you figure it out. Could we have said ‘this is on, this is off?’ Sure, we could have - but we’re not. Why? Because, (drop dead), that’s why. If you go look at the content outline for this exam, there’s nothing that highlights hey, this is a new paper (like used to be done when something was added), we’re calling this out so you notice.
Therefore, in the best interests of everyone, be aware there’s 2 major changes in the content:
- DELETE: NCCI Experience Rating Plan, NCCI Retro Rating Plan
- ADD: new GLM paper by Chalk et. al. - because, instead of doing a logical thing like splitting predictive modeling off into its own exam, or maybe even making an entire path that’s predictive modeling-related, we’re going to cram everything predictive modeling-related into an exam about advanced ratemaking that really should be focused on retros, experience mods, layer pricing, trending / detrending of limits, calculation of large loss loads and CAT loads, segmentation of data and resulting impacts on credibility for rate indications, good/bad complements of credibility, sensitivity-testing a rate indication (and what components most influence it, and why), …
1 Like
Thank you so much for this! I noticed that they added the new GLM paper but I missed the part where they took out the NCCI stuff.
As an aside, I completely agree with you on the focus of the exam. I have spent my entire career in pricing and most of the nuances that I deal with are not tested anywhere. I would love for an exam that touches on complements of credibility, different methods for calculating large loss loads, and how to balance the tradeoff between credibility and segmentation aside from “Actuarial Judgement”.
1 Like
I might even say “we should have an exam all around data quality, because it’s absolutely critical to our work, and we need to know how / when to infer values and how to do so correctly.” But that and other things are a rant for another time.
I kicked off studying over the weekend. I’m vaguely slow-walking it, but not really, but really trying to put more emphasis on the GLM crap because I hate it and I’ve (semi-)intentionally ignored it, and like when I tried doing Exam 5 when it was just ratemaking and they had the ISO PAP, ignoring a notable section of the syllabus tends to not work.
I’ll post notes / thoughts as I have time.
An important note for anyone who sees this.
TLDR: NCCI is still on the syllabus covered in the Fisher paper.
I checked the syllabus and confirmed with Howard Mahler that the NCCI topics are still on the exam covered in the Fisher paper. They removed the separate standalone NCCI papers which sometimes contradicted what is in the Fisher paper.
Is the mean excess loss e(x) the same as the loss greater than x given that a loss occurred?
Joe, e(x) is the expected loss excess of x given that the loss greater than x has occurred.
2 Likes
October 23, 2026. 12:30pm.
Let the cramming for this exam begin.
1 Like
Same here! Hopefully this is our sitting!
New exam study help from the CAS! It has 5 retired questions along with detailed answers. I didn’t look too closely at the questions but I do remember some of them and hopefully this will be the beginning of more resources for us candidates (and help us be better prepared for the exams)
1 Like