Approximate discrete-time schemes for statistics of diffusion processes. (English) Zbl 0704.62072

Let \(X=(X_ t,t\geq 0)\) be a diffusion type process which is a solution of the stochastic differential equation \[ (*)\quad dX_ t=b(X_ t,\theta)dt+\sigma dw_ t. \] Here \(w=(w_ t,t\geq 0)\) is a standard Wiener process, b(\(\cdot)\) is a given nonlinear function and \(\theta\), \(\sigma\) are unknown parameters. Let \(\Delta_ n\), \(n\geq 1\), be a sequence of (positive) numbers converging to zero or remaining constant. It is assumed that we have in our disposal the discrete observations \(\{X(k\Delta_ k)\), \(k=1,...,n\}\) obtained from a suitable first-order approximation for (*). Thus the general problem is to use these observations in finding estimators for \(\theta\), \(\sigma^ 2\) and study their asymptotic behavior. More precisely, the likelihood function of \(\{X(k\Delta_ k)\}\) is used as a contrast function in order to construct the contrast estimators \(\theta_ n\) and \(\sigma^ 2_ n.\)
Under suitable conditions the consistency (in \(L^ 2\)-sense) as well as the asymptotic normality are established. The case of a discretization with a constant sampling interval \(\Delta\) is analyzed in more detail in a separate section. Several related topics are discussed. Finally, one can conclude that the present paper contains the answers to interesting questions from statistics of diffusion processes by using discrete observations.
Reviewer: J.M.Stoyanov


62M05 Markov processes: estimation; hidden Markov models
60J60 Diffusion processes
62F12 Asymptotic properties of parametric estimators
60H10 Stochastic ordinary differential equations (aspects of stochastic analysis)
Full Text: DOI


[1] Bouc, R. and Pardoux, E.Asymptotic Analysis of PDES, 1984France: Publ., Math. Appl. Marseille. · Zbl 0574.60066
[2] Castelle Dacunha .D., Probability and statistics. 2 (1982)
[3] Castelle Dacunha .D., Stochastics 19 (1986) · Zbl 0626.62085
[4] DOI: 10.2307/3214063 · Zbl 0615.62109
[5] Dorogovtcev, Parameter estimation for stochastic processes (1982)
[6] Zmirou Florens .D., Note GBAS t 299 pp 595– (1984)
[7] Le breton A., Mathematical programming studies 5 pp 124– (1976)
[8] Breton Le A., Transactions of the 1th Prague Conference on Inf. Theory and Statistical decision-functions and Random processes pp 353– (1974)
[9] Mandl P., On dimensional Markov processes (1968) · Zbl 0179.47802
[10] >milshtein, Theory Prob, and App. 19 pp 557– (1974)
[11] Pardoux E., Ada Applicandal Math. 19 (1985)
[12] Simons Reed, Analysis of operators 4 (1976)
[13] Stoyanov. 1984. Problems of estiamtion in continuous-discrete models. Proc. of 7th Conf.on Prob. Theory (Brazov 1982). Bucharest IOSIFESCU Ed. 1984.
[14] Talay A., Resolution trajectorielle et analyse numerique des equaiton differentielles stochastiques (1983)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.