tag:blogger.com,1999:blog-5880610.post8852977645089441096..comments2020-11-29T02:07:43.911-05:00Comments on Information Processing: Central limit theorem and securitization: how to build a CDOSteve Hsuhttp://www.blogger.com/profile/02428333897272913660noreply@blogger.comBlogger12125tag:blogger.com,1999:blog-5880610.post-79105861294460509642008-11-19T12:26:00.000-05:002008-11-19T12:26:00.000-05:00Hi -- great comments. Unfortunately I am traveling...Hi -- great comments. Unfortunately I am traveling right now and can't make any substantive replies...<BR/><BR/>My treatment of CDO pricing is of course only the simplest of caricatures.Steve Hsuhttps://www.blogger.com/profile/02428333897272913660noreply@blogger.comtag:blogger.com,1999:blog-5880610.post-89276030892970521252008-11-19T02:13:00.000-05:002008-11-19T02:13:00.000-05:00The problem is clearly to find the joint distribut...The problem is clearly to find the joint distribution. (akin to the correlation, if those are gaussian)<BR/><BR/>Actually, the underlying idea that the problem is to find the distribution of the default is also wrong, as it might be completely (path?) dependant of other variable, like the equity market. so the joint distribution to find is even bigger.<BR/><BR/><BR/>In the end, we are completely Nicolashttps://www.blogger.com/profile/03434148024010872285noreply@blogger.comtag:blogger.com,1999:blog-5880610.post-67108699369357019592008-11-19T02:09:00.000-05:002008-11-19T02:09:00.000-05:00This comment has been removed by the author.Nicolashttps://www.blogger.com/profile/03434148024010872285noreply@blogger.comtag:blogger.com,1999:blog-5880610.post-41623723130442490012008-11-18T12:48:00.000-05:002008-11-18T12:48:00.000-05:00I don't think you need any thing as strong as the ...I don't think you need any thing as strong as the CLT. The weaker Chebyshev inequality should be enough.zarkov01https://www.blogger.com/profile/01062623377758910681noreply@blogger.comtag:blogger.com,1999:blog-5880610.post-73635657480948373422008-11-18T01:04:00.000-05:002008-11-18T01:04:00.000-05:00Diversification argument based on CLT is valid und...Diversification argument based on CLT is valid under assumptions of a finite variance. In the crisis underlying BBB's will have fat tail, more like power law, distributions. <BR/>In this situation diversification makes it worse.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5880610.post-1779364586526076662008-11-17T23:39:00.000-05:002008-11-17T23:39:00.000-05:00Okay, now for some behavioral economics: Spell out...Okay, now for some behavioral economics: Spell out how the probability of default becomes correlated.<BR/><BR/>Some answers: Borrower notices that other people are getting these no-money-down ARMs, and figures it must be a great deal. (See "bandwagon" and "sheep".) Or, large mortgage lender makes a habit of lending to people whose credit-worthiness is unknown or known to be poor. Or, economy Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5880610.post-18691353463664433312008-11-17T21:41:00.000-05:002008-11-17T21:41:00.000-05:00The principle of insurance is that many hands make...The principle of insurance is that many hands make light work, but that fails when carrying aquariums under waterfalls. The critical assumption is that mortgage loan defaults are independent. If that assumption is faulty, then there is a problem.William R. Hamblenhttps://www.blogger.com/profile/07796512452656973455noreply@blogger.comtag:blogger.com,1999:blog-5880610.post-58831368529669377782008-11-17T10:56:00.000-05:002008-11-17T10:56:00.000-05:00Hmm, the more I think about it, your CLT descripti...Hmm, the more I think about it, your CLT descriptions sounds more like a Law of Large Numbers description.<BR/><BR/>The CLT says to me, if I sample N things, and average them, I get a single average. If I execute this sampling M times, then the distribution of these averages is normally distributed.<BR/><BR/>In the case of the securities and tranches. Securitizing a set of mortgages is taking aAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-5880610.post-28094918610605667612008-11-17T06:47:00.000-05:002008-11-17T06:47:00.000-05:00"But we can improve our situation by making N iden..."But we can improve our situation by making N identical loans, assuming that the probabilities p of default in each case are uncorrelated -- i.e., truly independent of each other...."<BR/>why would you do that?<BR/>the easiest thing in the world to replicate is a portfolio of many N (large number ) loans : buy t-bills.jckhttps://www.blogger.com/profile/09902088903578116910noreply@blogger.comtag:blogger.com,1999:blog-5880610.post-11728345254435215682008-11-17T01:25:00.000-05:002008-11-17T01:25:00.000-05:00Thanks for explaining the fundamental structure of...Thanks for explaining the fundamental structure of CDOs.<BR/><BR/> I do have one question. You mentioned selling the first tranch = p* value of loan to somebody else.<BR/>Now buying something like this first tranch sounds extremely risky IF I knew what tranch I was getting. As a buyer what information do I get about this tranch before buying? Thanks for the answer in advance.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5880610.post-65238777786612076792008-11-16T20:16:00.000-05:002008-11-16T20:16:00.000-05:00Hmmm... I tried to tweak it but it makes sense to ...Hmmm... I tried to tweak it but it makes sense to me... perhaps I am missing some confusing aspect of the way I am expressing it?Steve Hsuhttps://www.blogger.com/profile/02428333897272913660noreply@blogger.comtag:blogger.com,1999:blog-5880610.post-35765259092010605442008-11-16T19:58:00.000-05:002008-11-16T19:58:00.000-05:00The central limit theorem tells us that, as N beco...<B>The central limit theorem tells us that, as N becomes large, the probability distribution of expected losses is governed by the normal or Gaussian distribution. The expected return is (r - p) times the total amount loaned, and, importantly, the variance of returns goes to zero as 1/N.</B><BR/><BR/>You say this in a very weird way for the point that you are trying to make.Anonymousnoreply@blogger.com