2,499 reputation
320
bio website researchgate.net/profile/…
location Vienna, Austria
age 32
visits member for 1 year, 10 months
seen 17 hours ago

Risk Manager at an Asset Management Company

External Lecturer at Vienna University of Technology


Jan
31
answered How does Vega of a call/put behave under the Black-Scholes model?
Jan
31
comment fair price for a call option
I agree to Anna: 1st you have to discount. My point of doubt is: it sound as if the investor estimates some probabilities and expects some return. But in reality you can not choose or wish probabilies and expected returns for options. If it is a text book example then you can take those probabilies and use the binomial pricing as Anna points out.
Jan
31
comment Need overlapping sample autocorrelation correction for calculating asset return correlations
I think it is not serious to model correlations of 20 year returns for the reasons I have explained. For statistical modelling you need some kind of staionarity. Im my mind this is not present in financial markets over such horizons.
Jan
30
comment fair price for a call option
You are aware of general option pricing theory with no-arbitrage arguments? How can we interpet the probabilities of the investor? The expected return that he wants to have? In reality the market does not care what the investor wants. Is this a problem of no-arbitrage option pricing or real option pricing?
Jan
30
comment Need overlapping sample autocorrelation correction for calculating asset return correlations
The paper that you refer to is certainly interesting, but I would not advise you to use the method if you are not familiar enough with time series analysis. For the second comment: you want to calculate asset return correlations for 20 years ... remember 1989 (opening of Eastern Europe), introduction of the Euro in Europe around the year 2000, dot.com bubble, 2008 crash. What do you want to model/predict/estimate based in 20 year periods? You have a bit more than 2 such periods from the end of WW2 1945 until 1989 ...
Jan
30
comment fair price for a call option
Use latex for formulas. Do not itemize with different symbols and so forth. It is not too much fun to read your question in the way it is now. Furthermore sentences like "What I did is... I calculate the using a tree graph the expected return that" are not correct and I am not willing to take time answering your question if you don't take time to formulate it.
Jan
30
comment fair price for a call option
Please improve the layout and typing of your question.
Jan
30
comment Portfolio Optimization : Shrinkage of Covariance Matrix when data is available
You don't want to use 10 years of data just because your universe contains a lot of stocks. It is debatable whether you want to use pre-2008 data for an optimization with a holding period of say 6 months from now.
Jan
30
comment Portfolio Optimization : Shrinkage of Covariance Matrix when data is available
No, there are 2 misunderstandings: shrinkage (or some other reduction technique) is necessary if you have more stocks than days AND (!) it is useful in general. For your second sentence: it must not depend on your sources. If you are a professional portflio manager then you have data for thousands of stocks. Obviously you might want to do some analysis on say 500 stocks using 1 year of look-back period. Then you need shrinkage. Count the stocks in big indices. You universe is not limited by online ressources but it is detemerined by the market that you want to cover and the look-back period.
Jan
30
comment Shortcomings of generalized Brownian motion for asset price modelling
I don't understand your second point - do you mean that they should be stochastic (like stochastic vol)? As Anna points out below the fact that in an SDE no jumps are modelled is one of the most important drawbacks in my mind.
Jan
30
answered Need overlapping sample autocorrelation correction for calculating asset return correlations
Jan
30
comment Portfolio Optimization : Shrinkage of Covariance Matrix when data is available
Just as a short comment: if the number of stocks is greater than the number of data points (e.g. days) then the sample covariance matrix is singular. Optimizers will reject such matrices as they allow for pathological results (if constraints do not prevent this). Furthermore if I have e.g. 250 days of data and 400 stocks (this something like $400*399/2$ correlations to estimate) how can I do this in a consistent way on 250 days - we should not believe too much in such an estimate.
Jan
29
comment Background required for the book by Brigo and Mercurio
The reason why such questions could be closed is because it is not useful to a broad audience. Your question is probably useful if you want to read this very book. Even then it is debatable. The standard textbooks on stochastic calculus are usually known.
Jan
29
comment Inferring signals in absence of sign of principal components (PCA)?
You are aware of the fact that you can flip all signs simultaneously? If not then I would advise you to study more details on how PCA works.
Jan
29
revised Understanding the VaR example on wikipedia
deleted 2 characters in body
Jan
27
comment Cross validation of a garch model
Maybe stats.stackexchange.com is a better place for this question.
Jan
22
comment Co-integration constraints of coint(X,Z) given coint(X,Y) and coint(Y,Z)?
Very good answer
Jan
22
comment Step-by-Step PCA algorithm (checking correctness without math packages)
Yes .. as far as I rememeber you had something like "largest eigenvalue" and "small variance". If you know that large eigenvalue measn large variance and vice-versa then it is fine.
Jan
22
comment Step-by-Step PCA algorithm (checking correctness without math packages)
You ask too much in one question. To answer a part: no: minimal variance is not related to the largest eigenvalue but rather too the smallest. And: inverting a matrix is not an estimator - what do you mean? Please rephrase the question. It starts with PCA and then goes to shrinkage. There is too much going on in this question
Jan
22
comment Co-integration constraints of coint(X,Z) given coint(X,Y) and coint(Y,Z)?
@aajajim You had the trick I was missing. Isn't it true (I read it, don't have the link, no time to prove it) that a linear combination of stationary processes is stationary? Then $\epsilon_t - \beta_1 \eta_t$ would be stationary and the proof would be finished (which I say as not being an expert in the field, maybe there are technical subtleties left).