934 reputation
220
bio website
location Vienna, Austria
age
visits member for 1 years, 9 months
seen 10 hours ago

Apr
15
comment how to extend lognormal model so that $\sigma$ is correlated to $\mu$?
@athos: I guess what you will end up with in the end will something like the "market price of risk" (the linear relationship you mentioned) and the notion of a risk-neutral measure. Please do not mix up correlation and simply a functional relationship. The correlation of deterministic quantities is always $0$. If you really want to pour time into this: This seems like a dead end to me.
Mar
31
comment Understanding the derivation of a ML-estimator
Could you clarify the first equation in (1)? There are some brackets missing I suppose? And why the notation $(\mathbb{1}^\prime \mathbb{1})^{-1}$? Isn't that just $1/T$?
Mar
21
comment Scaling of a transition matrix
@BrianB Nice!! Now that I read it, the Schur decomposition looks like the natural approach for this...
Mar
21
comment Scaling of a transition matrix
@Gerard Torrent OK I think I am missing something: Why should the eigenvalues of $D$ be nonnegative or even real? Why should all fractional powers exist (like 1/12 for example)?
Mar
19
comment What are the equation that gives hurst exponent of value >0.7 and <0.3?
I edited my answer.
Feb
19
comment What R-packages for SOCP problems are there?
@WilmerEHenaoH Why not? My problems are neither time-critical nor large scale...
Jan
21
comment Book on market microstructure
OK. Now you made me curious. Will check out your book...
Jan
13
comment Cointegration tests
The link appears to be dead.
Dec
18
comment Calculating and interpreting cumulative returns is R
@SBS Probably there is a require(TTR) missing. ROC() calculates discrete returns and you convert back with the exponential. You should correct that. Further more, if you model the price directly, try to reduce the sigma. You could also try something like seq(100,100.95,by=0.05) just to get started...
Dec
6
comment What is the right group of durations?
Do you mean the set of Ito Processes with constant coefficients? They will be proportional to $e^{R(t+W_t)-R^2t/2}$ of course and I think the "link" gets lost here because of the $R^2$-term. One more point: If you take $R_+^*$, it is only a group with the operation $*$, not with $+$. Semantically, when refering to duration, we measure the difference of time points. Of course, when using $+$, the group property and therefore the exponential as a homomorphism is lost. Thats why I asked about "group of durations"...
Dec
6
comment What is the right group of durations?
Can you define "group of durations"? you don't mean interest rate sinsitivities right? what is the group operation? And in your point 1, you have an mapping of $\mathbb{R}_+^*$ on what space?
Dec
6
comment Large (5K-10K) non positive definite (particularly near singular) covariance matrices and treatments for Cholesky decomposition
Ok, now we are getting close... :-) I couldn't agree more with what @John said. Keep in mind that parametrizing the curves and modelling the parameters will effectively result in a dimension reduction. If you wanted to use your cholesky decomposition to simulate random paths for example, one realization gives you all the changed curves. Using those, you can reprice ALL your thousands of futures contracts (from your curves) and thus get the price changes. There can be other issues though: You will see them when you get there ;-)
Dec
6
comment Single Most Important Fact about a Fund - Interview Question
Agree, this is a strange question. I think he just wanted to hear "performance". However, I hope the question will be closed.
Dec
5
comment Large (5K-10K) non positive definite (particularly near singular) covariance matrices and treatments for Cholesky decomposition
@acmh I said that. Basically, I suggested to use some kind of dimension reduction techniques. Instead of including all of those highly correlated returns you could take, say 1000 of them and use the returns of a corresponding index or sector instead. Probably, the derivations from the index returns would be about equal the estimation error. Roughly spoken, if you want to have more stable results, there is only one way: Try to reduce the estimation error by taking some bias. See also Richards shrinking suggestion. Thats also what you did by fiddling around with the Eigenvalues.
Dec
5
comment Large (5K-10K) non positive definite (particularly near singular) covariance matrices and treatments for Cholesky decomposition
IMHO, trying to estimate about 50 million parameters from only 1000 observations does not seem to be very wise. First of all, if there are linear dependencies, the matrix will be singular. If you fiddle around with the spectrum (by your method or shrinkage), this property will definitely be lost. You should try to reduce the dimension here. There are numerous ways to do this, all of them have their weaknesses. For a start, you could pool those returns with similar properties into asset classes and the problem becomes more feasible...
Nov
19
comment Are there any other standard rates term structure decomposition than PCA?
@user2763361 can you provide a paper/reference or more details?
Nov
18
comment Library for interactive financial charts
Did you look at GoogleVis? It comes as an R package too and should be quite easy to modify although I didn't put a lot of effort into it.developers.google.com/chart/interactive/docs/gallery
Oct
23
comment Examples of Spectral Risk Measures
@statquant Well, the aim is not to try a different spectrum $\phi$ but rather to identify the spectra of other well known risk measures. Brian B: I dont understand your comment about units. The units are irrelevant for a risk measure as far es the spectral property is concerned, right?
Oct
18
comment Is there a Bloomberg field for a bonds (upcoming) coupon dates?
Thats exactly what I was looking for, thank you.
Oct
17
comment Is there a Bloomberg field for a bonds (upcoming) coupon dates?
Yeah I found those but I didnt want to play around with calenders, weekends, public holidays and so forth. Nevertheless, there is the sad possibility that the field I am looking for does not exist. For now, I only upvoted your answer.