bostonbubble.com Forum Index bostonbubble.com
Boston Bubble - Boston Real Estate Analysis
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

SPONSORED LINKS

Advertise on Boston Bubble
Buyer brokers and motivated
sellers, reach potential buyers.
www.bostonbubble.com

YOUR AD HERE

 
Go to: Boston real estate bubble fact list with references
More Boston Bubble News...
DISCLAIMER: The information provided on this website and in the associated forums comes with ABSOLUTELY NO WARRANTY, expressed or implied. You assume all risk for your own use of the information provided as the accuracy of the information is in no way guaranteed. As always, cross check information that you would deem useful against multiple, reliable, independent resources. The opinions expressed belong to the individual authors and not necessarily to other parties.

Boston Bubble Wrap: The Real Story for MA - Apr 2011
Goto page Previous  1, 2
 
Post new topic   Reply to topic    bostonbubble.com Forum Index -> Greater Boston Real Estate & Beyond
View previous topic :: View next topic  
Author Message
admin
Site Admin


Joined: 14 Jul 2005
Posts: 1826
Location: Greater Boston

PostPosted: Mon Jun 06, 2011 8:20 pm GMT    Post subject: Reply with quote

BeaumontV wrote:
I think the admin is just trying to see what is more or less likely. But I do not want to speak for admin.


Yes, but there's a little more to it than that. I want to study how the probability distribution of the buy-vs-rent differential changes as various assumptions and inputs are changed. I also want to study which inputs exert the most influence on the outcome - I am guessing price, owning time, and inflation. The probability distributions of the inputs aren't necessarily known, so I want to try out a bunch of different "what if" scenarios to see where the shape of the distributions actually matters.

- admin
Back to top
View user's profile Send private message Send e-mail Visit poster's website
GenXer



Joined: 20 Feb 2009
Posts: 703

PostPosted: Wed Jun 08, 2011 12:15 pm GMT    Post subject: Reply with quote

BeaumontV wrote:
GenX

No one can predict the future. I think we can agree. what I think the admin wants to do is get odds on the future, which we can do. Given current events, the probabilities for some outcomes get better or worse.

Just because one cannot predict does not mean that calculating probabilities is futile. I think the admin is just trying to see what is more or less likely. But I do not want to speak for admin.


This is actually wrong, but it would take me more than a couple of pages to prove it. No, we can't measure future probabilities with any degree of certainty based on incomplete data such as futures.
Back to top
View user's profile Send private message
GenXer



Joined: 20 Feb 2009
Posts: 703

PostPosted: Wed Jun 08, 2011 12:20 pm GMT    Post subject: Reply with quote

To estimate odds you need to look at the fat tailed Case Schiller. Unfortunately, it will still not allow any reasonable predictions. The only thing that it can tell is the odds of big moves in the future. From my analysis, the odds of 8.3 sigma move is about 3:1000. This is about all you can get from case schiller. It is well known that you can't do anything with futures, and you can not get any type of estimate because you don't have enough data. To get the numbers I quoted above, I took Schiller's data going back to 1890. And from 3 months or 12 months of futures data you will get total garbage. You are welcome to try it, but it will be a total waste of time (because the data is not Gaussian, in a nutshell).

[/img]
Back to top
View user's profile Send private message
admin
Site Admin


Joined: 14 Jul 2005
Posts: 1826
Location: Greater Boston

PostPosted: Wed Jun 08, 2011 12:32 pm GMT    Post subject: Reply with quote

This isn't really relevant to what I'm doing (and I'd question your conclusions if it were). The futures only go out five years and I need to cover a longer time horizon. Also, it's Shiller (no 'c').

- admin
Back to top
View user's profile Send private message Send e-mail Visit poster's website
GenXer



Joined: 20 Feb 2009
Posts: 703

PostPosted: Wed Jun 08, 2011 12:33 pm GMT    Post subject: Reply with quote

If anybody wants to find out why we can't use futures data to make any type of reasonable probability estimates (unless they are of course, very wrong or right simply by accident), you are welcome to take the class I'm teaching at the Newton Community Education titled "Prudent Investment Management: Myths, Realities and Implications for Investors"

http://litovskymanagement.com/blog/newton-community-education-class.html

I'll be sure to challenge the entire worldview based on the Normal distribution (which is hopelessly wrong), and provide empirical ways to make sure that we are not fooled by randomness (part of which involves using statistics blindly without understanding where the limitations of statistical methods are - and there are limitations about which most practitioners are simply not aware of).
Back to top
View user's profile Send private message
admin
Site Admin


Joined: 14 Jul 2005
Posts: 1826
Location: Greater Boston

PostPosted: Wed Jun 08, 2011 12:51 pm GMT    Post subject: Reply with quote

GenXer wrote:
If anybody wants to find out why we can't use futures data to make any type of reasonable probability estimates (unless they are of course, very wrong or right simply by accident)...


And if anybody wants to see how accurate the futures actually have been for the Boston housing market (even though this is orthogonal to the current discussion) see: http://www.bostonbubble.com/latest.php?id=spcsi_bos_snapshot This is a straw man as it pertains to the current discussion, though.

- admin
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Beaumontv
Guest





PostPosted: Fri Jun 17, 2011 8:31 pm GMT    Post subject: Reply with quote

Ah, we have a Taleb disciple here. Putting. Futures aside for the moment, fat tails, non normal, Pareto distributions pose no problem for modeling. I happen to be at a quant conference right now and everyone here is well versed in dealing with non normal distributions. Monte carlo programs come pre loaded with many non Gaussian distributions.

Now as to the value of using futures, that depends.
Back to top
Display posts from previous:   
Post new topic   Reply to topic    bostonbubble.com Forum Index -> Greater Boston Real Estate & Beyond All times are GMT
Goto page Previous  1, 2
Page 2 of 2

 
Jump to:  
You can post new topics in this forum
You can reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Forum posts are owned by the original posters.
Forum boards are Copyright 2005 - present, bostonbubble.com.
Privacy policy in effect.
Powered by phpBB © 2001, 2005 phpBB Group