in

1/18/2008 - Beyond Monte Carlo Analysis: An Algorithmic Replacement for a Misunderstood Practice

Last post 01-28-2008 4:02 PM by Shawn Brayman. 6 replies.
Page 1 of 1 (7 items)
Sort Posts: Previous Next
  • 01-09-2008 3:20 PM

    1/18/2008 - Beyond Monte Carlo Analysis: An Algorithmic Replacement for a Misunderstood Practice

    Join the presenter, Shawn Brayman and other participants following Friday's Live Virtual Seminar (1/18,3:00 PM ET) as they share thoughts about the presentation.

    To participate in the conversation, you must first register and log-in to this site. The handouts for this session are available starting 24 hours prior to the session in the downloads area, under today's session date and title.

  • 01-18-2008 1:24 PM In reply to

    Re: 1/18/2008 - Beyond Monte Carlo Analysis: An Algorithmic Replacement for a Misunderstood Practice

    Was this conference recorded so I can relisten to some parts I didn't fully understand?

  • 01-18-2008 1:27 PM In reply to

    Re: 1/18/2008 - Beyond Monte Carlo Analysis: An Algorithmic Replacement for a Misunderstood Practice

    I missed the information about where the Reliability forecast is available.  Could you post the information.  Also, where is the paper available?  Does it/will it appear in the Journal of Financial Planning?

     

    Don Getinger

  • 01-18-2008 2:16 PM In reply to

    Re: 1/18/2008 - Beyond Monte Carlo Analysis: An Algorithmic Replacement for a Misunderstood Practice

    Hi Don

    The article is in the Journal of Financial Planning, December 2007 edition. It can be found online at http://www.fpanet.org/journal/articles/2007_Issues/jfp1207-art6.cfm

    The Reliability Forecast concept is obviously new and was first presented in September 2008 in Seattle at the FPA Conference and in the Journal in December 2007. As such I do not believe many people or firms have had the opportunity to digest the impact let alone implement something. PlanPlus has an implementation linked to a simple calculator, but do not expect something more elaborate before Q2-08. Hopefully other vendors will take the time to look at it as well.

    In the mean time, I think an important take away is that MCS and the Reliability Forecast are both "a function" of the assumptions. MCS is still a powerful tool as long as you understand what it does for you, and what it doesn't do!  

    I hope this helps 

    Shawn

  • 01-18-2008 2:53 PM In reply to

    RE: 1/18/2008 - Beyond Monte Carlo Analysis: An Algorithmic Replacement for a Misunderstood Practice

    Thanks for your reply.
     
    Sincerely,
     
    Don Gettinger
     

    Donald I. Gettinger, JD, CFP®, MA

    The Glowacki Group, LLC

     

    Come in for the answers. Stay for the questions.   

     

    11400 West Olympic Boulevard

    Suite 1500

    Los Angeles, CA 90064

     

    Phone: 310.473.0100

    Fax: 310.479.3145

    www.glowackigroup.com

     


    From: Shawn Brayman [mailto:bounce-239082@fpanet.org]
    Sent: Friday, January 18, 2008 1:18 PM
    To: Don Gettinger
    Subject: Re: [FPA_VLC_PostSession] 1/18/2008 - Beyond Monte Carlo Analysis: An Algorithmic Replacement for a Misunderstood Practice

    Hi Don

    The article is in the Journal of Financial Planning, December 2007 edition. It can be found online at http://www.fpanet.org/journal/articles/2007_Issues/jfp1207-art6.cfm

    The Reliability Forecast concept is obviously new and was first presented in September 2008 in Seattle at the FPA Conference and in the Journal in December 2007. As such I do not believe many people or firms have had the opportunity to digest the impact let alone implement something. PlanPlus has an implementation linked to a simple calculator, but do not expect something more elaborate before Q2-08. Hopefully other vendors will take the time to look at it as well.

    In the mean time, I think an important take away is that MCS and the Reliability Forecast are both "a function" of the assumptions. MCS is still a powerful tool as long as you understand what it does for you, and what it doesn't do!  

    I hope this helps 

    Shawn




  • 01-21-2008 10:43 AM In reply to

    Re: 1/18/2008 - Beyond Monte Carlo Analysis: An Algorithmic Replacement for a Misunderstood Practice

    This conference was recorded, the recording should be available sometime this week.

  • 01-28-2008 4:02 PM In reply to

    Re: 1/18/2008 - Beyond Monte Carlo Analysis: An Algorithmic Replacement for a Misunderstood Practice

    I wanted to share this question from Michael Ebal that was emailed to me directly.       
    Subject: Question on December JFP Article on MC

    Hello Shawn.  My name is Mike Ebel.  I read your article in December's Journal of Financial Planning titled, "Beyond Monte Carlo Analysis: An Algorithmic Replacement for a Misunderstood Practice".  Very interesting and enlightening. 
     
    But I have a question.  I notice that you made mention of Gobind Daryanani and his work on Sensitivity Simulations.  I read that work.  Also, I use MoneyGuidePro planning software that makes use of such sensitivity simulations through a  tool they coin, "Beyond Monte Carlo".  I am not strong mathmatically or statistically speaking, but I have enough of an understanding to muddle my way through.
     
    My question to you is: Isn't Daryanani's sensitivity analysis different from the typical monte carlo engines to which you refer in the article?  My understanding is that he feels this to be somewhat more accurate and certainly far less time consuming to come up with a result.  Or are you saying that we can group sensitivity analysis in with the problems/concerns that you outlined with the mainstream monte carlo engines?
     
    Thank you so much for your time and consideration.
     
    Michael J. Ebel, CFP®, CWS®

    Supervisor, Case Planning Operations

    Sigma Financial Corporation / Sammons Securities Company, LLC

    www.sigmafinancial.com

    www.sammonssecurities.com

    (P) 888-744-6264 x4528

    (F) 734-205-3529

     
    Hi Michael
     
    I am on the road at the moment and would like to review the article from Daryanani before responding, but based on my recollection you are quite right that he outlined a methodology he called Sensitivity Simulations that gave similar results as a Monte Carlo (or better when compared to only a small number of simulations 1000 or 2000) with only a subset of the calculations (60+/-). It was not an algorithmic solution but was definitely superior to a brute force Monte Carlo and I believed addressed issues like correlations between asset classes.
     
    So, my understanding is yes it is different, better in that in allows for correlations (not handled in MCS), faster and more sensitive when compared to small MCS runs.
     
    In respect to the last part of your question, the concerns I expressed were that any statistical methodology that applies a single Capital Market Assumption (mean and standard deviation) and "rolls up the results", is not in fact providing any true test of outlying returns - and I do not believe the methodology would matter - MCS, Sensitivity Simulations or my own Reliability Forecast. MCS can generate these results (albeit inefficiently) is if simulations were subgrouped and reported upon. If someone has implemented the Sensitivity Simulations, I cannot speak to what results they are either able to or are presenting.
     
    I hope this helps.
     
    Warm regards
     
    Shawn
     
     
Page 1 of 1 (7 items)
  Copyright © 2008 FPA All rights reserved Create/Modify Login Press Room Disclaimers Contact Us