8,144 reputation
930
bio website bachelierfinance.org
location United States
age 45
visits member for 3 years, 8 months
seen 11 hours ago

Worked in a lot of quant areas


Oct
17
comment What are the properties of the Expected Shortall measure when split in multiple time periods?
Agree...without an SDE for returns, you cannot relate ES over two time periods. That said it would be a very odd case that would have $ES_T > ES_{T+t}$
Oct
13
comment CDS Spread and Par Bond Yield Spread
To be fair, you've edited the question multiple times. In any case perhaps someone else will answer your question -- I am too unfamiliar with its origins and not interested in a research project.
Oct
13
comment CDS Spread and Par Bond Yield Spread
The mixup I perceived is that the nobody (so far as I ever met in either these markets and higher-rate regimes) ever reckoned that credit default swap (CDS) spread should approximate the risky par bond yield, mainly since (as you note) the risk-free rate needs to be added to it. (Including discrete coupons and working in integral space be $B,P$ introduces unneeded complexity to that aspect of the calculation). Perhaps you should cite the source of your folklore: it may be they are pretty dubious folk.
Sep
28
comment negative transition probabilities in the heston model
I can't really comment on the paper or the tree discretization in it, but trees seem like a pretty crazy way of solving the Heston PDE compared to implicit finite differences.
Sep
18
comment Solving Black-Scholes PDE using Laplace transform
It's perfectly legitimate to use the Laplace transform (and vonjd's linked paper does a fine job), but I've personally always preferred to solve the PDE by changing variables until the PDE turns into the standard diffusion equation.
Sep
12
comment What is Base- vs. Implied Correlation of a CDO tranche?
By "correctly price" I meant "matches the market price of". I'll update.
Aug
12
comment CDS - Accumulated Default Risk?
I agree, and will add a section on capital structure.
Aug
10
comment Discretization Schemes
I think your bias is reduced for every dimension in which you are using Milstein. But, if the two processes have "canceling" effects on the final result, you might be better off matching their discretization even if the match must be done at Euler level.
Jul
17
comment Arbitrage free implies complete market?
One often thinks of the cost of entering a position as the quantity $q$ times the difference $\nu$ between the offer (bid) and the midpoint. If the spread is $h$ then $\nu=h/2$, so I call it a half-spread.
Jul
1
comment What are the unfair order execution/routing advantages HFT firms apparently have?
It's probably worth adding that off-exchange (dark) pools, of which IEX is one, can make their own rules. IEX is not the only such entity violating price-time. Some dark pools subdivide their traffic into sub-pools according to client type, etc.
Jun
30
comment Self-financing and Black-Scholes-Merton formula
@Hansen unfortunately, as I say, ignoring the LCP is a "practioner thing" so references are rare. For example, I recall Monis doing this up through at least version 8.0, but to find out that fact I had to talk to the quants -- it was not in the documentation. Similarly for various internal investment bank pricing libraries. As to the policy iteration, I first saw a reference somewhere on the Wilmott forums around 2009(?), so you might start there.
Jun
27
comment Self-financing and Black-Scholes-Merton formula
PSOR tends to iterate the underlying Gauss-Siedel solver several times, especially if it hasn't been preconditioned with the "european" solution based on the naive exercise boundary (NEB). Many or most practitioners go to the extreme of completely ignoring the LCP and just applying the NEB, which introduces a fairly tiny error for most payoffs. A more efficient algorithm than PSOR to do this "right" is policy iteration.
Jun
27
comment Self-financing and Black-Scholes-Merton formula
Good document, though as a practical matter PSOR is not a good algorithm to solve the LCP.
May
27
comment Introducing credit risk to an already implemented interest rate model
You've basically already got it, though in most cases one assumes a deterministic term structure of credit spread, rather than making a stochastic model for it.
May
20
comment Create optimal portfolio by Treynor and Jensens Alpha
If you make your metric Sharpe-like, by including a measure of distributional width in the denominator, then there is of course a reward. If that measure is standard deviation, then the mathematics even works out to be nearly identical to Markowitz.
May
13
comment Are Futures exactly Delta One?
Matt's usage of the term <i>forward delta</i> is important here. RRG (and Swab Jat) seem to be forgetting that delta has a denominator. That is to say, when we speak of the delta of a contract, we sometimes need to be explicit about what security that delta refers to (this is particularly obvious for exotics like basket options).
May
12
comment Pricing binary options with kernel density estimation
If you price options off the real-world probabilities like that, you are going to lose your shirt.
May
1
comment Topological methods in finance
Interesting premise: a market metric based on Jones polynomial! I wonder if its behavior differs much from standard correlation metrics. Based on your username, the paper is your own -- it would be polite to mention that in your answer.
Apr
25
comment A question about pricing convertible bond with two different underlying assets
You'll get a better-quality answer if you are more specific about the terms and conditions (T&C) of the bond. In particular, if the holder chooses to convert, is the mixture of assets the conversion happens into determined by the holder, the issuer, or a formula?
Apr
17
comment Usage of Brownian Bridge?
For scrambling, a good place to start would be Hinckernell's Quasi-Monte Carlo methods and their randomizations.