2,499 reputation
320
bio website researchgate.net/profile/…
location Vienna, Austria
age 32
visits member for 1 year, 10 months
seen 6 hours ago

Risk Manager at an Asset Management Company

External Lecturer at Vienna University of Technology


Apr
14
comment Sampling problem in portfolio optimization
Some people try genetic/evolotionary algorithms you find hits on the web like this: citeseerx.ist.psu.edu/viewdoc/… For my taste branch-and-bound is the better choice.
Apr
14
comment Sampling problem in portfolio optimization
So, you say you want to "sample" the bonds. Maybe you want to pose a cardinality constraint meaning that you have a universe of $1000$ bonds and you want to do an optimization with certain constraints and get an optimal portfolio with $N$ bonds and $N \le K$ where $K$ is some maximal number of bonds that you want to hold. Is this interpretation right?
Apr
14
comment hedging with known volatility
Please tell us what your are asking.
Apr
11
comment HJM simulation problem
Could you please use latex for the formulas.
Apr
7
comment Why use implied volatility
@Ilya addressing your points: with useful I mean that BS is in any case useful as a tool to compare various options by their implied volatility (IV). High IV means rather expensive, small IV means rather cheap. BS with the correct IV (taken from the market) gives the price which is a tautology. If you use BS with the wrong IV then the price is wrong. For OTM options with short time to maturity BS is only able to get the price right with extremely large IVs. AD 2: yes it is equivalent. Again the IV smile/smirk is equivalent to the prices but on another scale.
Apr
3
comment Pricing an american style option on a bond future
Please improve the layout of your question.
Apr
3
comment Overview of robust/regularized portfolio selection
@John I thought that too. Basic concept but nice interpretation. Roncalli often presents such 'big picture'things.
Mar
31
comment Understanding the derivation of a ML-estimator
I could imagine that this fits better to stats.stackexchange
Mar
24
comment Geometric Brownian motion - Volatility Interpretation (in the drift term)
You meab $\exp(E[W_t]) \le E[\exp(W_t)]$ - right?
Mar
19
comment Which is the nearest town to London Gatwick
A joke ;) ... yes some quants might use Gatwick airport ;)
Mar
13
comment Normally Distributed Returns Become Leptokurtic Due to Compounding
In the two plots what is the standard deviation (sd) of the first random variable (is it $1$?) and what in the second? The axes and the sd should fit together. Or you use histograms.
Mar
13
comment Normally Distributed Returns Become Leptokurtic Due to Compounding
@jessica if $r$ is normally distributed with mean $0$ and variance $\sigma$ then $1+r$ is normally distributed with mean $1$ and variance $\sigma$. The distribution of $\prod_{i=1}^n (1+r_i)$ is not normal. If you look at log-returns then you can just sum up for accumulation over time. Then you stay in the log-return world.
Mar
13
comment Ex-Ante tracking error how to determine the look back period
What would make sense is to estimte TE ex-ante and then look at future active return. Similar to VaR back-testing.
Mar
13
comment Ex-Ante tracking error how to determine the look back period
Hi, do you want to compare ex-ante TE to ex-post TE? It depends on the purpose but this does not make too much sense too me. If you don't have a lot of trading then the numbers will be very close if you put in the same data, if you have a lot of trading then they will differ a lot.
Mar
5
comment How to backtest the VaR model?
An answer for a full Var model is another question, it depends on the asset class (bonds, equity, multi assets, HF). Post a question and you will get some references.
Mar
4
comment Can the duration of a bond be greater than Time to Maturity
Please tell us the exact example of this non-vanilla bond. Perpetual callable bonds for example get a "fake" maturity date on Bloomberg (something like 2045). Depending on the call schedule I can imagine that (effective) duration can be greater than this maturity.
Mar
3
comment Explanation or implementation of Ledoit-Wolf estimator (without math packages)
This is a very interesting question and very helpful that you provide code. Would you like to provide pseudo-code or something similar just to show the steps that your algorithm does? This would be somewhat clearer.
Feb
28
comment Sharpe Ratio, annualized monthly returns vs annual returns vs annual rolling returns?
I absolutely agree with @John : if you do not know $100\%$ what the interpretation is then don't do statistics on rolling returns. Furthermore: yes, use monthly or weekly returns. John, would you post this as answer - just to have this one answered.
Feb
21
comment why is the BNS model the way it is
In case you want to do Monte Carlo - I would work with $V$ in simulations and just plug in $\sqrt{V}$ for the simulation of $X$ - no Ito needed.
Feb
21
comment why is the BNS model the way it is
Of course the first derivative of a square-root is proportional to $1/\sqrt(...)$. But which application of Ito do you mean - an application to $f(X_t)$ or $f(V_t)$ for some $C2$ function $f$?