Actuarial Outpost
 
Go Back   Actuarial Outpost > Actuarial Discussion Forum > Software & Technology
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions


Reply
 
Thread Tools Display Modes
  #71  
Old 10-12-2011, 07:31 AM
JMO's Avatar
JMO JMO is offline
Carol Marler
Non-Actuary
 
Join Date: Sep 2001
Location: Back home again in Indiana
Studying for Nothing actuarial.
Posts: 36,246
Default

Quote:
Originally Posted by whisper View Post
I agree that testing an off-the-shelf PRNG from reputable sources would not be worth the effort.

Quote:
Would it really matter if they were repeating every million values?

Conceivably it could, if your cycle lined up perfectly with the PRNG repetition - but this is more than not unlikely.
If you are doing stochastic within stochastic, you need a very robust PRNG.

The practical question of "how many scenarios is enough?" also requires that adding more scenarios to your test does not result in repeating scenarios in the original set. Remember that for many applications, you need another random numbers for each year of the projection.
__________________
Carol Marler, "Just My Opinion"

Pluto is no longer a planet and I am no longer an actuary. Please take my opinions as non-actuarial.


My latest favorite quotes, updated Sep 18, 2017.

Spoiler:
I should keep these four permanently.
Quote:
Originally Posted by rekrap View Post
JMO is right
Quote:
Originally Posted by campbell View Post
I agree with JMO.
Quote:
Originally Posted by Westley View Post
And def agree w/ JMO.
Quote:
Originally Posted by MG View Post
This. And everything else JMO wrote.
MORE:
Quote:
Originally Posted by Dr T Non-Fan View Post
It would be rude to turn them down rudely. Try not to do that.
Quote:
Originally Posted by Woodrow View Post
Complementing your technical skills with social skills - whatever they are - will help your career. This is not a sign that the world is unjust.
Quote:
Originally Posted by Bro View Post
I recommend you get perspective.
Quote:
Originally Posted by Enough Exams Already View Post
Dude, you can't fail a personality test. It just isn't that kind of test.
Quote:
Originally Posted by Locrian View Post
I'm disappointed I don't get to do both.
Reply With Quote
  #72  
Old 10-13-2011, 12:24 PM
Gedankenexperiment Gedankenexperiment is offline
Member
 
Join Date: Dec 2010
Posts: 443
Default

Quote:
Originally Posted by JMO View Post
The practical question of "how many scenarios is enough?" also requires that adding more scenarios to your test does not result in repeating scenarios in the original set.
I assumed this question is answered by the output. If your distributions are smooth high resolution curves, then you're doing well. If not, you need more scenarios. If introducing more scenarios results in the same resolution, then it's time to sue Microsoft....I'm curious what happened with Mag's study.
Reply With Quote
  #73  
Old 08-12-2014, 11:12 AM
campbell's Avatar
campbell campbell is offline
Mary Pat Campbell
SOA AAA
 
Join Date: Nov 2003
Location: NY
Studying for duolingo and coursera
Favorite beer: Murphy's Irish Stout
Posts: 78,326
Blog Entries: 6
Default

Quote:
Originally Posted by campbell View Post



[As atomic mentions, there are some weird things that happen due to the floating point arithmetic in Excel... fixed point is obviously problematic, but even floating point has issues. I've been thinking of writing something up on floating point computation issues for CompAct, or maybe even one of the more technical actuarial journals. This can end up a real problem.... this is not necessarily getting OT with PRNGs, because some of the things I've come across on Excel discussion boards are weird results from RAND() because of precision issues. If you're using RAND() to feed into an infinite range distribution such as the gaussian, you're going to end up with NaN [not a number] if the PRNG gives you either 0 or 1].

The other thread reminds me: I did end up writing an article about floating point arithmetic (not specific to Excel)

https://www.soa.org/News-and-Publica...rithmetic.aspx
__________________
It's STUMP

LinkedIn Profile
Reply With Quote
  #74  
Old 03-30-2017, 11:06 AM
campbell's Avatar
campbell campbell is offline
Mary Pat Campbell
SOA AAA
 
Join Date: Nov 2003
Location: NY
Studying for duolingo and coursera
Favorite beer: Murphy's Irish Stout
Posts: 78,326
Blog Entries: 6
Default

https://medium.freecodecamp.com/a-br...s-9498737f5b6c

Quote:
A Brief History of Random Numbers
well, random & pseudo random. has some good stuff in there. check out the video
__________________
It's STUMP

LinkedIn Profile
Reply With Quote
Reply

Tags
data science, excel, predictive analytics, prngs, pseudorandom numbers, rand, random

Thread Tools
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 03:33 AM.


Powered by vBulletin®
Copyright ©2000 - 2017, 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 0.30838 seconds with 11 queries