search

jirotubuyaki / Jdmbs

An R Package for Monte Carlo Option Pricing Algorithm for Jump Diffusion Models with Correlational Companies
GNU General Public License v3.0
28 stars 6 forks source link

Mixed-Exponential Jump Diffusion #1

Open stevenlvkai opened 5 years ago

stevenlvkai commented 5 years ago

Why you use a uniform random variable multiplying a normal random variable as Mixed-Exponential Jump Diffusion? J[j] <- J[j] + runif(1,min=0,max=1) * rnorm(1, mean = 0, sd = 1)

jirotubuyaki commented 5 years ago

Thank you for pointing out the issue. I implemented it for random jump size. But even if I remove "runif(1,min=0,max=1)" , Random jumps will occur. Do you think this codes have problems?

stevenlvkai commented 5 years ago

Thank you for pointing out the issue. I implemented it for random jump size. But even if I remove "runif(1,min=0,max=1)" , Random jumps will occur. Do you think this codes have problems?

No, I don't think this is a problem. I just don't understand why using a uniform distribution here. It seems like the uniform distribution just shrinks the amplitude of the jump. As you said, rnorm() has already provided a random generator. So, why we need a uniform variable to adjust it? Is there any paper mentions this method?

jirotubuyaki commented 5 years ago

I mainly refer "Stochastic Calculus for Finance II: Continuous-Time Models (Springer Finance)". But this book is not mention jump size and I don't refer any academic papers.

stevenlvkai commented 5 years ago

I mainly refer "Stochastic Calculus for Finance II: Continuous-Time Models (Springer Finance)". But this book is not mention jump size and I don't refer any academic papers.

Ok, understand. Thank you very much. I really appreciate your work. Maybe you can mention a little more detail about $Yi$ in the mixed-exponential jump-diffusion model in your vignette.

jirotubuyaki commented 5 years ago

I will better my repository and vignette. Thank you very much.