Serving the Quantitative Finance Community

Search found 11287 matches

by Alan
January 12th, 2006, 3:16 pm
Forum: Forum and Website Bugs and Suggestions
Topic: Several screens of 'debugging info"
Replies: 1
Views: 125112

Several screens of 'debugging info"

<t>The cold fusion server is feeding me several screenfuls of "debugging information"which appear at the bottom of my browser (IE 6.0... SP2) on the visit to any forumpage. There are tables labelled 'execution times', 'sql queries', 'scope variables', etc.This just started today (thurs, Jan 12).rega...
by Alan
January 12th, 2006, 2:12 am
Forum: Book And Research Paper Forum
Topic: Books of functional analysis
Replies: 27
Views: 194777

Books of functional analysis

For the Peter Lax fans:Interview
by Alan
January 8th, 2006, 4:06 am
Forum: General Forum
Topic: equity-bond correlation
Replies: 28
Views: 128793

equity-bond correlation

<t>QuoteOriginally posted by: erstwhileEquity-bond correlation used to always be positive (except during a flight to quality as in 1987), and the conventional wisdom was that this made sense for XYZ economic fundamental reason. There was even a guy whose main job was to trade "the mean-reversion of ...
by Alan
January 7th, 2006, 2:46 pm
Forum: Student Forum
Topic: Standard error of a volatility estimate
Replies: 4
Views: 126960

Standard error of a volatility estimate

Hi Vito,Well, you've almost solved it. Given a distribution for a random variable X, what is thedistribution for sqrt(X)? Once you have the distribution, take N large.regards,
by Alan
January 6th, 2006, 4:56 pm
Forum: Student Forum
Topic: Standard error of a volatility estimate
Replies: 4
Views: 126960

Standard error of a volatility estimate

<t>This is a standard topic in statistics called the "sampling distribution of the variance".If you are a student, you could learn a lot by seeing how far you can get withoutlooking it up. Assume s = Sum (x_i)^2/N (or N-1), where the x_i are independent draws from a normaldistribution. Try to figure...
by Alan
January 3rd, 2006, 3:10 pm
Forum: Student Forum
Topic: Riddle me this (Wright-Fisher diffusion)
Replies: 7
Views: 125543

Riddle me this (Wright-Fisher diffusion)

Hi Peter,No, you can freely choose any integration constant (away from 0 or 1). Whatever you choose, C1 changes to keep the final answer the same.regards,
by Alan
January 1st, 2006, 4:39 pm
Forum: Student Forum
Topic: Riddle me this (Wright-Fisher diffusion)
Replies: 7
Views: 125543

Riddle me this (Wright-Fisher diffusion)

<t>Hi Peter,Ok, well first to answer your edited question: when kappa is not zero,you can't count on finding the long-run mean by setting the drift term tozero. (Nor can you expect to find a general SDE solver.)Karlin and Taylor or many other books will give the following prescription.Your SDE has t...
by Alan
December 30th, 2005, 3:11 pm
Forum: Student Forum
Topic: Riddle me this (Wright-Fisher diffusion)
Replies: 7
Views: 125543

Riddle me this (Wright-Fisher diffusion)

Given -any- stationary diffusion, it is a standard matterto work out the stationary density and from this you canobtain the lon-run mean. See Karlin and Taylor (a 2nd course)for details.good luck and happy new year,alan
by Alan
December 14th, 2005, 11:50 pm
Forum: Technical Forum
Topic: Is Energy a Giffen good?
Replies: 8
Views: 128554

Is Energy a Giffen good?

<r>Everything I know about Giffen goods I just learned at Wikipedia. They attribute this idea about gasoline (and you said it wasn't your own)to Sasha Abramsky in this article:<URL url="http://www.thenation.com/doc/20051017/abramskyIs">http://www.thenation.com/doc/20051017/abramskyIs</URL> that wher...
by Alan
December 14th, 2005, 10:59 pm
Forum: Technical Forum
Topic: Blind pricing
Replies: 3
Views: 127156

Blind pricing

<t>Well, as a first thought: if this is -really- a practical question,then there is a real company with a real history about to do an IPO.This company probably has an internal option plan that has beenrunning for quite some time. So, one step would be to investigate all the parameters that they use ...
by Alan
December 2nd, 2005, 7:53 pm
Forum: Student Forum
Topic: How to construct the risk less portfolio with SABR model?
Replies: 9
Views: 129247

How to construct the risk less portfolio with SABR model?

<t>QuoteOriginally posted by: LapinSorry my mistake.When I calculte dP, I have to take into accoun the fact that U depends on V and S.Sorry for that.But is the way to solve it still valid?RegardsSure, write Stcohastic44 for the details.If you do it that way, you can indeed contruct a riskless hedgin...
by Alan
December 2nd, 2005, 7:10 pm
Forum: Technical Forum
Topic: Does the short term smile vanish in SV models?
Replies: 7
Views: 129514

Does the short term smile vanish in SV models?

<t>QuoteOriginally posted by: PutorCallConsider the simplest SV model:dS_t = sigma_t dW_tdsigma_t = dZ_twhere W and Z are independent standard Brownian motionsand S_0 = sigma_0 = 1. Then we know that Bachelier implied vol of some finite maturity option is symmetric in strike K about K=1, not flat, a...
by Alan
December 1st, 2005, 5:15 pm
Forum: Technical Forum
Topic: guys, please help to solve this equation
Replies: 3
Views: 129080

guys, please help to solve this equation

<t>You're welcome.Most of Mathematica's documentation is online.There are two arbitrary constants C[1] and C[2].The result may look quite obscure.To help explain the notation, the slot # is a formal parameter; replaceit with some parameter, say x. Then, inside the InverseFunction[...],you have a fun...
by Alan
December 1st, 2005, 3:05 pm
Forum: Student Forum
Topic: How to construct the risk less portfolio with SABR model?
Replies: 9
Views: 129247

How to construct the risk less portfolio with SABR model?

<t>You can't make it riskless if you assume the volatility risk is unhedgable.You can zero out the equity risk only. The net effect of doing this is aa risk-adjusted 2 factor system (S,V) where (i) dS = r S dS + Sqrt(V) S dB(t)(ii) dV = b(V) dt + a(V) dW(t). The noise terms Sqrt(V) S dB(t) and a(V) ...