<t>Thank you very much for your response. I computed the average BS implied vol for each moneyness category (ITM, ATM, OTM) for different range of maturities. I find weird results that the term structure is increase when is or close to ATM, decreasing for OTM, increasing for slight ITM and flat for ...

<t>I saw some places say B-S implied volatility increases with longer days to maturity under a certain strike price/moneyness. However, I also saw other places report a decreasing function. Can anyone explain to me if it is increasing or decreasing over days to maturity? or maybe the answer is mixed...

<t>Hi Kielers, are you trying to calibrate a jump diffusion model with both price and vol jumps? yes, I agree it is hard. I am also working on this model ...Hi Ben, totally agree with you. I found using option prices make it take a very long time for the optimization, plus the unexpected broken code...

<t>Hi Kielers, thank you for providing thoughts in this post. Personally I never think the more complicated the model is, the better. However, I think it is important to have a model to contain the crucial properties that we observe from real data, while we try to keep the model as simple as possibl...

Hi Alan and Ben, thank you very much for your help! Sorry that I didn't check the post more frequently. I added a specified small tolerate in the integration code, hopefully it can improve. I will follow your suggestions and do it more. Thanks

<t>Hi, I wrote code evaluate option prices under a jump-diffusion model with both price jump and volatiity jump. The last part of the pricing code is to do an integration to get option analytical price. For many parameter sets I tried, it works perfectly. But when I run an optimization to search a p...

hi SPV205, thank you for your response. do you mean it shouldn't take 40 second to calculate option prices when using numerical ODE? so it is possible that my code is wrong? do you know how long it usually takes?

<r>Thank you very much for your response. This is the package I use. It is written in R software. Thank you for your more comment in advance.<URL url="http://rgm2.lab.nig.ac.jp/RGM2/func.php?rd_id=deSolve:zvode"><LINK_TEXT text="http://rgm2.lab.nig.ac.jp/RGM2/func.php ... olve:zvode">http://rgm2.lab...

<t>I need to solve A and B both as functions of t. it has two equations like:B_t=?B^2+?B+?+?/(?+?B), A_t=?B+?. where A_t and B_t are derivative of A and B over t. ? are coefficients with complex numbers. I am using software to solve ODE numerically to calculate the option price. My ODE has no analyt...

hi, can anyone share some experiences on solving numerical ODE for complex numbers on matlab? thanks

<t>Hi, everyoneI find that "zvode" in the package of "deSolve" in R can compute the numerical ODE that has complex numbers. Has anyone here used it before? how does it work? how is the accuracy? speed? How does it work compared to ode45 in matlab? is there a better one in R to numerically solve ODE?...

<t>QuoteOriginally posted by: frenchXQuoteOriginally posted by: TinManOut of interest, how many parameters are in that model?Too many in my opinioninitial volvol of volvol mean reversionlong term volvol correlationasset jump distributionvol jump distribution (in fact it's a 2 d distribution because ...

<t>QuoteOriginally posted by: frolloosi haven't managed to correlate jumps well, if you do, let me know (i know there is a paper somewhere on correlating VG processes though.)Thanks! I am doing my phd thesis. One goal of this paper is to propose an econometric method to estimate this model. I will u...

QuoteOriginally posted by: frenchXI will have a closer look at your calculus on Friday evening . Today and tomorrow I'm really busy. SVSCJ model, oh dear Thank you very much!!! I am deriving some stuff of this model for the model calibration. My intuition told me it might work well

<t>QuoteOriginally posted by: frolloosdoesn't local vol have an infinite number of parameters? but it's still used.Hi, Thanks for the comment. The point volatility Vt is a latent process. so in this sense, yes, it has infinite number of parameters. I treat it as a latent process to calibrate the mod...