Graham
About
- Username
- Graham
- Joined
- Visits
- 1,722
- Last Active
- Roles
- Administrator
Comments
-
Yes, the error in the unbiased estimator may average out to 0 over many years but you could have a situation where the true value of the parameter is 50 but over many observations, you see predictions anywhere from 20 to 80 (distributed randomly) Fo…
-
Ok so unless you let me know, I'll assume you don't need any further explanation.
-
You don't. Your method will work.
-
This was question maybe a little unfair. It was different from the text examples of the adjusted pure premium approach because we aren't given the current relativities. Normally we would use the current relativities to adjust the exposures. Since we…
-
That's fine. You'll get the same answer either way.
-
Actually, the data is sort of a mixture between annual and quarterly. According to the column heading "Year Ending Quater", each row in the table represents 12 months of data, but it's "rolling years". That means that at the end …
-
If I understand your question correctly, it shouldn't matter what we let the areas be as long as they are all correct relative to each other. They are dividing by 0.25 as you highlighted because the total area is 0.25 and the values in parentheses (…
-
You're welcome!
-
The solution is based on this formula for current rate level factor (CRLF): CRLF = CRL / ARL where CRL stands for Current Rate Level and ARL stands for Average Rate Level. You know CRL from part (a): CRL = 1.03763 The hard part is calculating ARL. H…
-
They were trying to trick you here. When they say "full-term" premium, it doesn't mean "annual" premium. Full-term refers to the actual policy term, which for policy C is 6 months. In other words, the policyholder paid $1,000 for…
-
Sample answers 1 & 2 give the same result because the actual severities for AY 2014 and 2015 were irrelevant. As you noted, the development pattern did not change so whether or not you multiply the reported losses by 0.75 for AY 2014 & 2015,…
-
That certainly is an exhaustive solution! And now that the exam is given in Excel, you can actually solve it like that on the exam. I want to say yes to your question about the aggregate formula. I only hesitate because who knows what kind of strang…
-
Yes, a lack of data means method 1 probably can't be used for any entity, but a self-insured entity is much more likely to have this problem because the only data they have is their own. A normal insurance company probably has many insureds and lots…
-
I'm not sure which question you're referencing: QB? There's only 1 old exam question from Spring 2014 (Q3) but this doesn't seem to be what you're asking about.
-
I would need to see your calculations to better comment on your proposed method. A couple of points however: The ending unearned exposures on Dec 31, 2017 are not relevant because they will be fully earned by Dec 31, 2018 anyway. The first sample an…
-
I think you meant to reference a different question. This doesn't seem to be about Spring 2013 Q3.
-
You can't calculate it like that. You don't have enough information. (You need the paid counts.) When a claim is paid, it is settled for the correct amount and the over-reserving is reversed in the next row entry in the reported triangle. So you als…
-
If I understand correctly what you did, if you rebase the indicated relativities before calculating the rate change (and doing the off-balancing) then the rate change for the base class would always be 0.0% which wouldn't make sense. (It isn't that …
-
An example for the first bullet point could be the item in parentheses above: Both new and renewal policies count as 1 exposure. That means an exposure-based method could produce the same expense dollars for both new and renewal policies but new pol…
-
(See above for part (b). For part (c) where you have to comment on the reasonableness of the ULAE estimate, here's how I thought through it: The standard assumption for the classical ULAE method is that 50% of the cost is incurred when the claim is …
-
Here are a few general comments before I get to the actual solution: The examiners report did say this was a difficult problem and that candidates scored poorly with many not even attempting it. There are usually 1 or 2 problems like that on every e…
-
You're welcome!
-
In the Werner text, "loss cost" means "losses / exposures". It's synonymous with "pure premium". (So "loss costs" and "losses" are different things.)
-
Sorry, I didn't state it properly. I meant to say the amount earned for all policies written in the second half of the year would average to 50%. (The only policies that would be fully earned would be the ones written on July 1.)
-
Step 1: Calculate written exposures by dividing the given written policies by 2. This is because the written policies are 6-month policies, so the total written exposures is half the written policies. Step 2: Apply this formula: Earned Exposures = 0…
-
No, the loss ratio method is for calculating an indicated rate change. That's different from calculating indicated relativities for territories. To calculate relativities, we need the loss costs for each territory so we can compare them. Then territ…
-
There was a trick in this problem. They tell you the retention ratio was 77% when the company wrote 6-month policies and are essentially asking you to calculate (or rather estimate) the retention ratio after they switched to annual policies. If you …
-
It's quite a coincidence that this was the only example where your method didn't work. To see why it didn't work, let's look at the contribution to the CY 2022 incurred loss from claim C. You wrote 154, but the correct value is -57 because: The onl…
-
You're welcome!
-
The concept here is that once you've calculated the proposed base rate and realized that the (AOI<100K) x (Territory 1) combination does not meet the minimum premium requirement, you have to do 2 extra steps: Calculate the premium shortfall for t…