Actuarial Outpost
 
Go Back   Actuarial Outpost > Search Forums
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions


Fill in a brief DW Simpson Registration Form
to be contacted when new jobs meet your criteria.


Showing results 1 to 25 of 300
Search took 1.25 seconds.
Search: Posts Made By: Heywood J
Forum: Property - Casualty / General Insurance 06-20-2018, 11:07 AM
Replies: 11
Views: 563
Posted By Heywood J
The statistical answer is that you reprice your...

The statistical answer is that you reprice your holdout policies with the new rating plan and compare to how you would've done versus the old rating plan.
Forum: Property - Casualty / General Insurance 06-19-2018, 06:10 PM
Replies: 11
Views: 563
Posted By Heywood J
If you use Ginis to measure your lift, the middle...

If you use Ginis to measure your lift, the middle improvement will be reflected. I don't think anyone actually measures lift merely by comparing the last bucket to the first bucket, that's just an...
Forum: Property - Casualty / General Insurance 06-19-2018, 02:46 PM
Replies: 11
Views: 563
Posted By Heywood J
No. Lift measures the spread of actual losses...

No. Lift measures the spread of actual losses when ordered by predicted losses. It does not measure the spread of predicted losses when ordered by predicted losses.
Forum: Exam 5 (new) - Ratemaking/Reserving 05-06-2018, 07:29 PM
Replies: 134
Views: 12,585
Posted By Heywood J
Seriously, I think that would be a more practical...

Seriously, I think that would be a more practical way to go. I would definitely want to practice on the same laptop that I would take my test on, and I don't want my proctors be the ones ensuring...
Forum: Exam 5 (new) - Ratemaking/Reserving 05-06-2018, 07:20 PM
Replies: 134
Views: 12,585
Posted By Heywood J
I think that you can get a laptop for $500 that...

I think that you can get a laptop for $500 that would be more than adequate for this task. Getting them to and from the testing center seems like a bigger problem, especially in NYC.

The most...
Forum: Property - Casualty / General Insurance 03-30-2018, 12:30 PM
Replies: 16
Views: 1,865
Posted By Heywood J
What is the point of modeling claims with no...

What is the point of modeling claims with no payment? Whatever the frequency of them, wouldn't you have to multiply it by expected severity of zero when calculating the rate?
Forum: Property - Casualty / General Insurance 03-19-2018, 05:01 PM
Replies: 2
Views: 556
Posted By Heywood J
What are you trying to accomplish by doing this...

What are you trying to accomplish by doing this comparison?
Forum: Property - Casualty / General Insurance 03-09-2018, 01:16 PM
Replies: 17
Views: 1,969
Posted By Heywood J
Perhaps we are thinking of different things when...

Perhaps we are thinking of different things when we talk about calibration. If your model is ranking well but missing with some identifiable pattern, then you can adjust its predictions with the...
Forum: Property - Casualty / General Insurance 03-09-2018, 12:33 PM
Replies: 12
Views: 977
Posted By Heywood J
You offset for the log of relativity, the other...

You offset for the log of relativity, the other way is just wrong. Offset is a predictive variable that you force to have a coefficient of one. It is logical to force a coefficient of one on the...
Forum: Property - Casualty / General Insurance 03-09-2018, 11:29 AM
Replies: 12
Views: 977
Posted By Heywood J
Yes, subtract 2000 or earliest year, otherwise...

Yes, subtract 2000 or earliest year, otherwise the silly scale of some numbers is going to be distracting.
Forum: Property - Casualty / General Insurance 03-08-2018, 10:16 PM
Replies: 17
Views: 1,969
Posted By Heywood J
I think it's a bigger mistake to look at a bunch...

I think it's a bigger mistake to look at a bunch of different statistics, when most of them are potentially misleading, irrelevant, or redundant. Gini is the only statistic you need to evaluate the...
Forum: Careers - Employment 03-01-2018, 12:09 PM
Replies: 6
Views: 755
Posted By Heywood J
Or the y-axis label, for that matter. Either one...

Or the y-axis label, for that matter. Either one will give you an answer, though not necessary the same one.
Forum: Careers - Employment 03-01-2018, 12:03 PM
Replies: 28
Views: 1,506
Posted By Heywood J
You can go wrong in both directions. Actuarial...

You can go wrong in both directions. Actuarial field truly requires a mix of practical and academic knowledge. Relying too much on one to the detriment of the other can be dangerous in either case.
Forum: Careers - Employment 02-18-2018, 01:42 PM
Replies: 33
Views: 4,018
Posted By Heywood J
I have not read this article in detail, but one...

I have not read this article in detail, but one thing in general to keep in mind is that the vast majority of such articles are highly intellectually dishonest. There are countless ways to evaluate...
Forum: Careers - Employment 02-07-2018, 01:19 PM
Replies: 124
Views: 13,859
Posted By Heywood J
If we're talking about data science in insurance,...

If we're talking about data science in insurance, being clever and experienced is much more important than whether you have a graduate degree. There are a lot of PhDs in the industry who are either...
Forum: General Actuarial 12-11-2017, 01:42 PM
Replies: 149
Views: 11,882
Posted By Heywood J
They evolved that way because that's what works. ...

They evolved that way because that's what works. Rote is king when time pressure is the limiting factor.
Forum: General Actuarial 12-11-2017, 11:17 AM
Replies: 149
Views: 11,882
Posted By Heywood J
Older actuaries who are now in management should...

Older actuaries who are now in management should still have some skills in the area. They're the ones setting the agenda, so they can't make use of data science if they have little practical idea...
Forum: General Actuarial 12-10-2017, 03:18 PM
Replies: 149
Views: 11,882
Posted By Heywood J
To be fair to ewood, data science is indeed a...

To be fair to ewood, data science is indeed a poorly defined term, and that's a shortcoming acknowledged by data scientists themselves first and foremost. People may think of different roles without...
Forum: General Actuarial 12-08-2017, 03:32 PM
Replies: 149
Views: 11,882
Posted By Heywood J
Data scientists at their core are big data...

Data scientists at their core are big data statisticians, whereas classical statisticians (and actuaries to a large extent) are small data statisticians. The size of dataset affects pretty much...
Forum: General Actuarial 12-08-2017, 01:42 PM
Replies: 149
Views: 11,882
Posted By Heywood J
If you work a lot with data scientists, you'll...

If you work a lot with data scientists, you'll probably find quite a few that are both very smart, and who dropped out of the exam process. Who knows how many other people didn't even stay in the...
Forum: General Actuarial 12-07-2017, 06:03 PM
Replies: 149
Views: 11,882
Posted By Heywood J
As a counter-argument, there are a number of...

As a counter-argument, there are a number of reasons to consider the opinion of mere students as more than just worthless whining.


They're actually on the business end of the process right now....
Forum: General Actuarial 12-07-2017, 12:55 PM
Replies: 149
Views: 11,882
Posted By Heywood J
That's neither here nor there. The comments...

That's neither here nor there. The comments about exams are not the issue, the comments about the comments are.

Treating every comment about the exam process as just useless whining that goes...
Forum: General Actuarial 12-06-2017, 09:09 PM
Replies: 149
Views: 11,882
Posted By Heywood J
That's the kind of refreshing attitude that keeps...

That's the kind of refreshing attitude that keeps actuarial science at the forefront of innovation.
Forum: Property - Casualty / General Insurance 11-26-2017, 03:27 PM
Replies: 2,058
Views: 253,589
Posted By Heywood J
Skepticism can be useful, but uninformed or...

Skepticism can be useful, but uninformed or misinformed skepticism is grating. Image recognition has undergone a revolution over the last 5-10 years. Autonomous vehicle developers are making use of...
Forum: Property - Casualty / General Insurance 11-26-2017, 02:33 AM
Replies: 2,058
Views: 253,589
Posted By Heywood J
People that think we're close understand that...

People that think we're close understand that autonomous cars process visual cues in a far more advanced way than any production car with technology package sold today.
Showing results 1 to 25 of 300

 

All times are GMT -4. The time now is 07:36 AM.


Powered by vBulletin®
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
*PLEASE NOTE: Posts are not checked for accuracy, and do not
represent the views of the Actuarial Outpost or its sponsors.
Page generated in 0.34182 seconds with 7 queries