Actuarial Outpost
 
Go Back   Actuarial Outpost > Exams - Please Limit Discussion to Exam-Related Topics > SoA > Modules 6-8
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions

2017 ACTUARIAL SALARY SURVEYS
Contact DW Simpson for a Personalized Salary Survey

Reply
 
Thread Tools Search this Thread Display Modes
  #221  
Old 09-14-2018, 11:35 AM
timmbo1987 timmbo1987 is offline
Member
SOA
 
Join Date: Mar 2014
Studying for Global Conquest
College: Ball State University, Alumni, M.A.
Favorite beer: Coffee and more Coffee
Posts: 108
Default

DNMMR - Tasks 2 & 4. I guess I am at a loss at how to re-tackle Task 2. Task 4 will follow the risk/return requirements and metrics that I use in Task 2, so I am not really worried about that question. But, I am starting to get a little flustered. This is my second DNMMR for the Final Assessment.

So, my approach to Task 2 was to initially set a maximum level for volatility (which essentially limits the amount invested in risky assets, or equities), then my goal was to find the lowest possible CTE at the 95% confidence level. I ran something like 20 scenario, found the scenario that matched my risk/return requirements, and then I used that asset mix percentage (which was 25% Treasuries, 65% Bonds, and 10% Equities). Then, I used the mean for that given scenario to be the annual contribution amount.

I thought that was a good approach to the question, but for some reason, and for the second time, the graders did not agree.

Any thoughts as to how one may improve this approach or how you may approach this problem?

Should I set the Equity Percentage first (say 10%), run dozens of scenarios, and then find the one that has the lowest coefficient of variation? I am thinking of maybe using a ratio for this attempt? Ultimately we want the volatility for the asset mix to be low as we want the fund to remain solvent. That is the primary, fundamental purpose of this fund.
__________________
P/1, FM/2, MFE/3F, MLC/3L, STAM , PA
VEE Econ , VEE Stats , VEE Fin
FAP 1, 2, 3, 4, 5, I.A.,6, 7, 8, F.A.
Reply With Quote
  #222  
Old 12-11-2018, 04:45 PM
actuary28 actuary28 is offline
Member
SOA
 
Join Date: Sep 2016
Posts: 44
Default

I am not sure if anyone is still following this discussion but here goes:

I see many people especially Sir Isaac saying that he minimized mean/CTE. This would mean that we are maximizing CTE. What is the logic behind that? My gut feeling tells me hes on to something but I can't seem to figure it out.
Thanks!
Reply With Quote
  #223  
Old 01-12-2019, 11:52 AM
Sir Issac's Avatar
Sir Issac Sir Issac is offline
Member
SOA
 
Join Date: Jun 2014
Posts: 409
Default

Quote:
Originally Posted by actuary28 View Post
I am not sure if anyone is still following this discussion but here goes:

I see many people especially Sir Isaac saying that he minimized mean/CTE. This would mean that we are maximizing CTE. What is the logic behind that? My gut feeling tells me hes on to something but I can't seem to figure it out.
Thanks!

I think I mentioned it before but the basic idea was if two portfolios have a similar cte the one with the lower mean has better risk/return tradeoff. So you can think about it as as minimizing the mean for portfolios with similar CTE risk.

Higher CTE implies higher risk, with higher risk comes higher potential for higher return. So by maximizing the CTE and minimizing the mean I am getting best risk/return trade-off.

Also I posted this, I’m quoting it in case you or someone else finds it helpful:

If I were to do it again I personally would incorporate Sharpe Ratio instead. I would maximize the following ratio under a certain requirement for equity allocations (higher equities bring more volatility year over year and CDEF cannot afford short term losses):

(Average cost per employee from 100% treasuries - average cost per employee from a certain portfolio mix) / standard deviation

(Some people told me they're using CTE in this ratio instead of mean which also makes sense to me since you would be using the mean from worst case scenarios opposed to the general mean)

By maximizing this ratio I would be attaining the lowest possible cost per employee when comparing the cost per employee attained from the risk free rate while taking volatility into account. It's similar to the sharpe ratio example I gave but using cost per employee instead.

Please don’t use it if you don’t understand it. Attempt to understand it first

Last edited by Sir Issac; 01-12-2019 at 11:56 AM..
Reply With Quote
  #224  
Old 02-09-2019, 08:28 PM
mistersunnyd mistersunnyd is offline
Member
SOA
 
Join Date: Aug 2016
Studying for how to find a job
Posts: 174
Default

When they say "available risk/return metrics", do they mean separate risk AND return metrics or metrics that measure both risk and return? Also, what do they mean by "available"? Was something given?
Reply With Quote
  #225  
Old 02-09-2019, 09:39 PM
mistersunnyd mistersunnyd is offline
Member
SOA
 
Join Date: Aug 2016
Studying for how to find a job
Posts: 174
Default

Oh blimey Harry it's in the frickin spreadsheet!!! Can't believe I spent three hours on this...
Reply With Quote
  #226  
Old 02-21-2019, 05:52 PM
actuary28 actuary28 is offline
Member
SOA
 
Join Date: Sep 2016
Posts: 44
Default

Quote:
Originally Posted by Sir Issac View Post
I think I mentioned it before but the basic idea was if two portfolios have a similar cte the one with the lower mean has better risk/return tradeoff. So you can think about it as as minimizing the mean for portfolios with similar CTE risk.

Higher CTE implies higher risk, with higher risk comes higher potential for higher return. So by maximizing the CTE and minimizing the mean I am getting best risk/return trade-off.

Also I posted this, Iím quoting it in case you or someone else finds it helpful:

If I were to do it again I personally would incorporate Sharpe Ratio instead. I would maximize the following ratio under a certain requirement for equity allocations (higher equities bring more volatility year over year and CDEF cannot afford short term losses):

(Average cost per employee from 100% treasuries - average cost per employee from a certain portfolio mix) / standard deviation

(Some people told me they're using CTE in this ratio instead of mean which also makes sense to me since you would be using the mean from worst case scenarios opposed to the general mean)

By maximizing this ratio I would be attaining the lowest possible cost per employee when comparing the cost per employee attained from the risk free rate while taking volatility into account. It's similar to the sharpe ratio example I gave but using cost per employee instead.

Please donít use it if you donít understand it. Attempt to understand it first

Hi Sir Isaac,

I actually understood it. Thank you! I wouldn't have used it if I didn't understand it because I wouldn't have been able to explain it. thank you so much!!

Can I ask you one more thing? When they say to discuss the Pros and Cons, does that mean we need to list the pros and cons of them separately? For example like list the pros and cons for the mean, var, maximum....or list the pros and cons of a combination of risk/return metric? You explanation from before makes me think its the latter since the mean, var, maximum...are just expressions of return and not the actual risk/return metric.

Thank you again! Really appreciate it!!
Reply With Quote
  #227  
Old 02-22-2019, 04:01 PM
BabyHorse15 BabyHorse15 is offline
Member
CAS SOA
 
Join Date: Apr 2015
Studying for Exam PA
Posts: 47
Default

Quote:
Originally Posted by actuary28 View Post
Can I ask you one more thing? When they say to discuss the Pros and Cons, does that mean we need to list the pros and cons of them separately? For example like list the pros and cons for the mean, var, maximum....or list the pros and cons of a combination of risk/return metric? You explanation from before makes me think its the latter since the mean, var, maximum...are just expressions of return and not the actual risk/return metric.
FWIW, I listed pros and cons for each individually (mean, VaR, CTE, max/min, etc) and for the risk/return metric I selected. Listing pros and cons individually helped me in the explanation of why I selected my risk/return metric.
__________________
Reply With Quote
  #228  
Old 02-22-2019, 04:30 PM
actuary28 actuary28 is offline
Member
SOA
 
Join Date: Sep 2016
Posts: 44
Default

Quote:
Originally Posted by BabyHorse15 View Post
FWIW, I listed pros and cons for each individually (mean, VaR, CTE, max/min, etc) and for the risk/return metric I selected. Listing pros and cons individually helped me in the explanation of why I selected my risk/return metric.

Hi BabyHorse15,

Thank you so much! Really appreciate it!
Reply With Quote
  #229  
Old 03-05-2019, 06:12 PM
Sir Issac's Avatar
Sir Issac Sir Issac is offline
Member
SOA
 
Join Date: Jun 2014
Posts: 409
Default

Quote:
Originally Posted by actuary28 View Post
Hi Sir Isaac,

I actually understood it. Thank you! I wouldn't have used it if I didn't understand it because I wouldn't have been able to explain it. thank you so much!!

Can I ask you one more thing? When they say to discuss the Pros and Cons, does that mean we need to list the pros and cons of them separately? For example like list the pros and cons for the mean, var, maximum....or list the pros and cons of a combination of risk/return metric? You explanation from before makes me think its the latter since the mean, var, maximum...are just expressions of return and not the actual risk/return metric.

Thank you again! Really appreciate it!!

I listed the pros and cons separately.
Reply With Quote
  #230  
Old 03-21-2019, 05:00 PM
Sir Issac's Avatar
Sir Issac Sir Issac is offline
Member
SOA
 
Join Date: Jun 2014
Posts: 409
Default

Someone sent me a private question and I thought I'd be helpful to put the question here with the answer in case someone relates to the question.

Quote:
Hi SirIssac, I've been reading your posts regarding Task 2, and while I agree with why you want to minimize the mean, I'm a little confused as to why you want to maximize CTE. I'm leaning towards minimizing CTE so that we minimize risk. You said that if we maximize CTE, "with higher risk comes higher potential for higher return", but if we're already minimizing the mean which is essentially maximizing return, I don't see why we would still want to maximize CTE? It sounds like you're taking a high risk, high return approach instead of a high return to risk approach. Please let me know if I'm misunderstanding what you said. Thanks in advance.
Answer:

Quote:
I minimized mean/cte and I'll explain why I chose that:

Scenario 1: If portfolio A and portfolio B have CTE = 900 for example but portfolio A has mean = 500 and portfolio B has mean = 600.

What is a better risk/return trade-off? I would say it's portfolio A, since the mean is lower meaning the mean cost per employee is lower. If I wanted to maximize this ratio, I would have to recommend portfolio B which doesn't make sense since I would be choosing a portfolio with a higher mean of cost per employee which isn't rational. I'm sure we both agree on this right?

I initially thought about it that way then realized that by minimizing mean/cte, mathematically you're either minimizing the mean or maximizing CTE or both. If you take an approach where you're minimizing CTE, then you're increasing the mean/CTE ratio, which counters the outcome of the example I just provided.

Scenario 2: Now let's say portfolio A and B both have mean of 500 but portfolio A has CTE = 900 and portfolio B has CTE = 950.

What is a better risk/return trade-off? I would say this is a harder question to answer. Your average risk is still the same but the tail risks are different. You can argue on of the two points:

1. choosing CTE = 950 is higher because there is a higher potential for return and since the mean is 500 in both cases, it's better to choose 500 / 950 (minimizing mean/cte)

2. You can argue that since the tail risk is lower at CTE = 900 you should choose 500 / 900 (I think this is the approach you're thinking about). This would maximize the mean/CTE, which goes against scenario 1.

Scenario 3: Now let's say portfolio A has mean = 500 and CTE = 950 and portfolio B has mean = 550 and CTE = 900.

What do you choose in this case? It's not clear which one to choose. If you only want to minimize the mean, you would have to choose portfolio A but the CTE is higher in portfolio A and that is more tail risk. Do you choose portfolio B? It's less tail risk but it's higher average risk.

So if your risk return metric is to minimize mean/cte, you would have to go with portfolio A. If your risk/return metric is to maximize mean/cte you would choose portfolio B.

So it comes down to defining why you chose the risk/return metric you did. In my case I think minimizing mean/cte is better because I would get a better risk/return trade-off by choosing the lower mean of 500 and higher CTE of 950 (higher potential risk at the tail, but also higher potential return). But if I choose mean = 550 I would be giving up the low risk of having a lower mean to have a better tail risk of 900 (So I would be trying to avoid a more extreme worst outcome than if I chose 950 but also at the sacrifice of having higher mean risk).

This is how I thought about it. I'm sure someone can argue the other way if they explain why they prefer their risk tolerance to be the other way.
Reply With Quote
Reply

Tags
cdef, fa task 2, final assesment

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


All times are GMT -4. The time now is 06:07 PM.


Powered by vBulletin®
Copyright ©2000 - 2019, 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 0.17870 seconds with 9 queries