<t>Hi guys,I managed to get access to a very nice level 2 order book database for my master's thesis and now I am thinking about which relationships in order book data are worth looking at.Apart from some obvious things (like a correlation between the bid-ask spreads of the 10 quotes) what kind of r...

the first problem I see is that you only have one path of your TA-strategy. maybe a null-set event ruined the performance of your strategy during that time.

<t>hello everyone,I have to use d-vines with garch marginals for order book data in my master's thesis.does anyone have recommendations for a garch extension that allows non-equally spaced data?I haven't got any experience with the problem of non-equally spaced data as in my class on linear time ser...

<t>suggestion:I leave the pde of the stock price the way it is and I set the brownian motion of the volatility process to be the sum of rho times the stock brownian motion + sqrt(1-rho^2) times a second independent brownian motion.then the two pdes will be driven by a two independent brownian motion...

<t>hi,why can you use girsanov for the heston pde. In the books I know the theorem is introduced for processes driven by an n-dimensional brownian motions only (which before were defined to be a vector of n independent 1-dim brownian motions).I am sure that there is a way to use it but I have not be...

<t>I worked on trend following strategies during an internship that had surprisingly good properties and I am interested in "formalizing" some trading rules in my master's thesis.The professor I want to write my thesis with is doing research on markov switching models and would be interested if I us...

<t>I have the following homework in an applied lecture which unfortunately has very little information about the theoretical background."Find the VAR of an asset swap (long corporate bond and short swap (pay fixed rate) -remember term structure of spreads!"Until know, I only have the theoretical bac...

I am not sure whether I understood your question but you are missing 1/2 it is(u-(sigma^2)/2)t

thank you very much! I hope people don't rely too much on those conventions. Especially the annualization of a one-day VaR appears kind of meaningless to me.

<t>I will try to be more specific:If I assume that my returns are i.i.d. normally distributed because I believe in markowitz and I use the square-root rule on monthly vols and then extrapolate to yearly returnsI should get a significant error w.r.t. to the model's "real" vol.The motivation to use th...

<t>Hi guys,I am a bit confused about scaling volatility depending on whether we assume i.i.d. normal log returns driven by a geometric brownian motion or i.i.d. simple returns in the markowitz framework.Assume now we are in the markowitz world and we have i.i.d. normal simple returns with return RTh...

<t>I chose the structuring guys.it's hard to judge for me who has the reall experts.the company I am doing the internship with is quite new and they are the portfoliomanagers for the new goldman/merril/morgan stanley european ETF platform (source).to me that sounded quite appealing. but of course th...

<t>hello guys,I am currently enrolled in a math. finance master's at tu munich and I have two internship offers with quant. asset managers.one would be about return forecastings. either on the use of option pricing models to forecast stock returns or on the cds spread as a factorthe other internship...

for short time periods you can assume the return to be 0 as it is usually relatively small compared to volatility.there is an example in Hull options futures and other derivatives

<t>Actually I don't see your pointwe have Var(f(X))~ (f'(E[X])^2 *Var(X)which in the case of X= 1+R and f(x) = x^nyields n^2* (1+E[R]])^2(n-1) * Var[R]then I assumed that E[R] = 0 ...and then we have growth with n^2*Var[R]the assumption of E[R]= 0 also removes the compounding effect to summarize:I a...