99

An approximate maximum likelihood estimator of drift parameters in a multidimensional diffusion model

Glasnik Matematicki - Serija III (Glas. Mat.), 2023
Abstract

For a fixed TT and k2k \geq 2, a kk-dimensional vector stochastic differential equation dXt=μ(Xt,θ)dt+ν(Xt)dWt,dX_t=\mu(X_t, \theta)dt+\nu(X_t)dW_t, is studied over a time interval [0,T][0,T]. Vector of drift parameters θ\theta is unknown. The dependence in θ\theta is in general nonlinear. We prove that the difference between approximate maximum likelihood estimator of the drift parameter θnθn,T\overline{\theta}_n\equiv \overline{\theta}_{n,T} obtained from discrete observations (XiΔn,0in)(X_{i\Delta_n}, 0 \leq i \leq n) and maximum likelihood estimator θ^θ^T\hat{\theta}\equiv \hat{\theta}_T obtained from continuous observations (Xt,0tT)(X_t, 0\leq t\leq T), when Δn=T/n\Delta_n=T/n tends to zero, converges stably in law to the mixed normal random vector with covariance matrix that depends on θ^\hat{\theta} and on path (Xt,0tT)(X_t, 0 \leq t\leq T). The uniform ellipticity of diffusion matrix S(x)=ν(x)ν(x)TS(x)=\nu(x)\nu(x)^T emerges as the main assumption on the diffusion coefficient function.

View on arXiv
Comments on this paper