233 reputation
16
bio website
location
age
visits member for 1 year
seen Feb 17 at 3:45

Apr
10
awarded  Yearling
Dec
15
awarded  Citizen Patrol
Dec
15
awarded  Informed
Dec
13
answered EuroDollar vs FRA
Dec
8
answered How to compute interest rate futures spread ratio?
Aug
18
comment VaR Calculation - Covariance matrix is not positive semidefinite
A covariance matrix has to be positive semi-definite (and symmetric). If it is not then it does not qualify as a covariance matrix. Yes you can calculate the VaR from the portfolio time series or you can construct the covariance matrix from the asset time series (it will be positive semi-definite if done correctly) and calculate the portfolio VaR from that.
Aug
18
answered VaR Calculation - Covariance matrix is not positive semidefinite
Jul
15
awarded  Teacher
Jul
15
answered Am I reading this correctly? probability way too small with BS model
Apr
23
awarded  Scholar
Apr
23
awarded  Supporter
Apr
23
accepted Auto-correlation of GBM
Apr
23
comment Auto-correlation of GBM
It seems your solution is correct. I will accept it as an answer to my question. Thanks.
Apr
11
comment Auto-correlation of GBM
You are here calculating the covariance between two independent GBMs. Indeed, substituting back $\gamma=\mu-\sigma^2/2$ into your formula gives $Cov(S_t,S_r) = e^{\mu(t+r)}\left(e^{\sigma^2r}-1\right)$. In particular without drift, $\mu=0$, we have $Cov_{\mu=0}(S_t,S_r) = e^{\sigma^2r}-1$, which is independent on $t$ and is just the variance of a GBM. What I am looking for is the auto-correlation of a single GBM trajectory: $S_0\rightarrow S_t \rightarrow S_{t^\prime}$.
Apr
10
awarded  Student
Apr
10
asked Auto-correlation of GBM