×

Maximum contrast estimation for diffusion processes from discrete observations. (English) Zbl 0721.62082

The author considers an one-dimensional diffusion process \(X=(X_ t\), \(t\geq 0)\) defined as the solution of the stochastic differential equation \[ dX_ t=b(\theta,X_ t)dt+\epsilon dW_ t,\quad t\geq 0,\quad X_ 0=x. \] Here W is a standard Wiener process, \(\epsilon\) is a positive “small” parameter, b(\(\theta\),x) is a given nonlinear function and \(\theta\) is an unknown parameter, \(\theta \in \Theta \subset {\mathbb{R}}^ k\). The problem is to find the maximum contrast estimators (m.c.e.) of \(\theta\) on the basis of the observations of X at times \(\Delta\),2\(\Delta\),..., i.e. by using the so-called equally spaced data.
Some necessary definitions and preliminary statements are given. The relationship between the m.c.e. and the m.l.e. is studied when simultaneously \(\epsilon\) and \(\Delta\) tend to zero. It is quite interesting that the m.c.e., whose existence is established, obeys different asymptotic behavior as \(\epsilon\to 0\) and \(\Delta\to 0\), and properties like consistency and asymptotic normality primarily depend on the connection between \(\epsilon\) and \(\Delta\). All the statements are proved in detail and illustrated well by nice examples. Related topics are also discussed.
Let me note that the present paper as well as papers of other French stochasticians (to mention A. Le Breton, R. Azencott, D. Dancunha- Castelle, M. Duflo, D. Florens-Zmirou, C. Denieu, C. Laredo) are an essential contribution to the statistics of stochastic processes.

MSC:

62M05 Markov processes: estimation; hidden Markov models
62F12 Asymptotic properties of parametric estimators
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Azencott R., Geometrie differentielle stochastique 921 pp 237– (1982)
[2] Dacunha Castelle. D, Probability et statistiques 2, Problemes a temps mobile (1983) · Zbl 0535.62004
[3] Dactjhha Castelle. d, Stochastics 19 pp 263– (1986) · Zbl 0626.62085 · doi:10.1080/17442508608833428
[4] Flobens zmirou, d, Statistics 20 pp 547– (1989) · Zbl 0704.62072 · doi:10.1080/02331888908802205
[5] Gewom Oatalot. V, These Universite Paris-Sud (1987)
[6] Gunon, Cataiiot, V and Laredo, C. 1986.Exhaustivite asymptotique de differentes observations partielles d’une diffusion, 303Paris: C.R.A.S.
[7] Ibragimov I.A., Statistical estimation, Asymptotic theory (1981) · Zbl 0467.62026
[8] Ikeda N.S., Stochastic differential equations and diffusion processes (1981) · Zbl 0495.60005
[9] Ktjtoyants Yu, Parameter estimation for stochastic processes (1984)
[10] Le breton A., Mathematical Programming Study 5 pp 124– (1976)
[11] Liptseb R.S., Statistics of random- processes I, General theory (1977)
[12] Stoyanov J., Problems of estimation in continuous-discrete stochastic models (1982)
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. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.