English

Parameter estimation in CKLS model by continuous observations

Statistics Theory 2021-05-31 v1 Probability Statistics Theory

Abstract

We consider a stochastic differential equation of the form drt=(abrt)dt+σrtβdWtdr_t = (a - b r_t) dt + \sigma r_t^\beta dW_t, where aa, bb and σ\sigma are positive constants, β(12,1)\beta\in(\frac12,1). We study the estimation of an unknown drift parameter (a,b)(a,b) by continuous observations of a sample path {rt,t[0,T]}\{r_t, t \in [0,T]\}. We prove the strong consistency and asymptotic normality of the maximum likelihood estimator. We propose another strongly consistent estimator, which generalizes an estimator proposed in Dehtiar et al. (2021) for β=12\beta=\frac12. The identification of the diffusion parameters σ\sigma and β\beta is discussed as well.

Keywords

Cite

@article{arxiv.2105.13724,
  title  = {Parameter estimation in CKLS model by continuous observations},
  author = {Yuliya Mishura and Kostiantyn Ralchenko and Olena Dehtiar},
  journal= {arXiv preprint arXiv:2105.13724},
  year   = {2021}
}

Comments

13 pages

R2 v1 2026-06-24T02:33:57.868Z