2007 Annual Financial Engineering Case Competition

[Uncle Max]

I just accidentally found out that this event will be hosted today, April 13th 2007. It's sad that we didn't manage to go there. I hope things will change next year

2007 Graduate Business School Financial Engineering Competition Hosted by Tepper School of Business, Carnegie Mellon University and Lehman Brothers Sponsored by Appaloosa Management, Lehman Brothers, and PRMIA Institute Our Tepper students will again be traveling to New York for the annual Financial Engineering Competition where they will compete with students from the top graduate programs in quantitative finance.The competition will be held on Friday April 13th at the New York offices of Lehman Brothers, 745, 7th Avenue at 49th Street. A private welcoming reception will be held on Thursday April 12th from 6:00 to 8:00 pm. at Tonic Bar, 2nd floor, 727 Seventh Avenue (Between W. 48th & W. 49th). All Tepper alumni who work at Lehman Brothers are invited to attend this reception and are asked to RSVP in advance. On Friday, April 13, 2007 6:30 pm to 8:30 pm there will also be a Closing Receptiuon at the Lehman Brothers Executive Dining Room on the 32nd Floor (Club I). To attend, alumni should email Carissa at: carissa.ng@lehman.com

 

[alain]

Could we enter next year? At least we have an email now that we can use.

 

[Bridgett]

If I understand correctly, according to what I learned from Greg (our risk instructor), this is kinda exclusive to a narrowly defined group (and we are still considered as outsiders). Greg was trying to get us in, but I guess more aggregated and continuous efforts are needed? I imagine any connection in Lehman's would be helpful for us to tackle it next year... anyone?

 

[Uncle Max]

As for that email address, it's for the Lehman alumni RSVP purpose only.

I think they mean CMU alumni

 

[alain]

If I understand correctly, according to what I learned from Greg (our risk instructor), this is kinda exclusive to a narrowly defined group (and we are still considered as outsiders). Greg was trying to get us in, but I guess more aggregated and continuous efforts are needed? I imagine any connection in Lehman's would be helpful for us to tackle it next year... anyone?

does anybody has a connection with Lehman Brothers? I think this is a perfect platform to showcase our skills.

 

[John]

it is worth revisiting the 2005 financial engineering competition case, which focused on HEL ABS, now that this asset class is receiving so much attention:

Winterland Asset Management: Asset-Backed Securitzation of Subprime Home Equity Loans (2005)

• A hedge fund wants to invest in high-yielding credit derivatives but wants to hedge the risk of a collapse in house-prices.

http://bertha.tepper.cmu.edu/fec/winterland.pdf

 

[Bridgett]

it is worth revisiting the 2005 financial engineering competition case, which focused on HEL ABS, now that this asset class is receiving so much attention:

http://bertha.tepper.cmu.edu/fec/winterland.pdf

Indeed interesting... thanks for the pointer.

 

[John]

Indeed interesting... thanks for the pointer.

anytime!

Here's the Lehman HEL paper quoted on the case sheet in case someone is interested.

 

[samratbatth]

Solution to Winter asset Management Case

Hey guys,

Did anyone did actually solve this case as an exercise?
Just asking.

Thanks,

 

[kipucamayoc]

Hi,
any clu for the solution?

 

[kipucamayoc]

clue*

 

[John]

Hi,
any clu for the solution?

I am willing to work with you if you are interested.

 

[kipucamayoc]

Sure, let me know if I understand this :we have 4 risk factors, variations of interest rates, volatility, bankruptcy, and interest rate changes and their effects.
I was thinkin to Do a one factor model such s vasichek ( OR SOME OTHER DISRETE) for each and calibrate. Combine the interest rate changes, and the interest rate variability with some form of instantaneous correlation, then try to combine that with the volatility model with those( yes with instantaneous correlations) and do the same with the defaults. If we use continuous then instantaneous correlations, if discrete then find a way to include the correlations ( will not be instantaneous, because of intervals).

what do you think?