scientific article; zbMATH DE number 1810239
Let \((P_t,\sigma_t)\) be a diffusion process given by \[ dP_t = \sigma_t dW_{1t}, \quad d\sigma_t^2 = (\omega+\theta\sigma_t^2) dt+ \alpha\sigma_t^2 dW_{2t}. \] The authors propose an approximation of the process \((P_t,\sigma_t)\) based on a discrete time sequence \((P_{kh},\sigma_{kh})\) satisfying \[ P_{kh}=P_{(k-1)h} + \sigma_{(k-1)h}\varepsilon_{kh}, \] \[ \sigma_{kh}^2 = \sigma_{(k-1)h}^2+\omega h+(\alpha h-1)\sigma_{(k-1)h}^2 + \beta_h\sigma_{(k-1)h}^2\varepsilon_{(k-1)h}^2, \quad\varepsilon_{kh}\sim \text{i.i.d. } N(0,h). \] It is shown that the approximation converges weakly to \((P_t, \sigma_t)\). Pseudo-maximum likelihood method is used for estimating parameters of \((P_{kh},\sigma_{kh})\). This allows to avoid estimating parameters from extensive simulations.
- Estimating a class of diffusions from discrete observations via approximate maximum likelihood method
- An Approximate Innovation Method For The Estimation Of Diffusion Processes From Discrete Data
- A new estimating function for discretely sampled diffusions
- Estimation of diffusion parameters for discretely observed diffusion processes
- A two-stage realized volatility approach to estimation of diffusion processes with discrete data
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3148089)
