scientific article; zbMATH DE number 1154475
The paper presents an exposition of the theory of stochastic differential equations more general than Itô ones. The topics are the following: stochastic integrals, linear and nonlinear equation existence and uniqueness, dependence on the initial data, Markovian property, generators, convergence to a stationary regime, multiparametrical equations and SPDEs. The author is mostly interested in higher order equations, the interest of which is justified via the example of a particle motion in mechanics. The important feature of such (higher order) systems is their degeneracy. First of all, in Chapter 2 (Chapter 1 being an introduction) a stochastic integral \(\int_0^tf(s) dX_s\) is introduced for some class \(\mathcal O\) of processes \(f(s)\) and for \(X_s\) which satisfies the \(L^{\varphi_{1},\varphi_{2}}\)-boundedness principle: \[ E\varphi_2\Biggl(\int_T f dX\Biggr) \leq K \int_{T\times\Omega} \varphi_1(f) d\alpha, \quad f\in {\mathcal O}, \] for some measure \(\alpha(dt\times d\omega)\). Then in Chapter 3 a linear theory is exposed for equations of the type \[ dX(t) = (\alpha(t)X(t) + \alpha_0(t)) dt + \sum_{i=1}^k (\beta_i(t)X(t) + \gamma_i(t)) dB_i(t) \] with \(B_i\) \(L^{2,2}\)-bounded (i.e. \(L^{\varphi_{1},\varphi_{2}}\)-bounded with \(\varphi_i(x) = x^2\)) and nonrandom \(\alpha, \beta, \gamma\). Explicit formulas are presented. In particular, the Wiener process is \(L^{2,2}\)-bounded. It is also noticed with reference to the author's book ``Stochastic processes: General theory (1995; Zbl 0841.60002) that the Skorokhod integral is also included in \(L^{2,2}\)-theory. Moreover, the Markov property is shown for \(X\) with independent increments. The semigroups associated with the equation are discussed as well in this case. Further, in Chapter 4 the nonlinear theory is exposed again with the main interest in higher order equations. Existence and uniqueness are established under Lipschitz conditions on coefficients. The Markovian property is discussed briefly. The semi-group property leads naturally to the study of generators. Some general assertion is stated in Chapter 4 which then is discussed in more details for the Brownian case. In Chapters 6 and 7 the truncated moments for a small time are calculated for this latter case which means that one knows a generator on smooth functions, due to the theory of Markov processes. Then Doob's one-dimensional theorem is exposed which gives the generator in terms of the first two truncated moments. In Chapter 8 some comparison theorem due to \textit{J. A. Goldstein} [Nagoya Math. J. 36, 27-63 (1969; Zbl 0186.49901)] is formulated. A version of a (weak) ergodic theorem is stated in the case when the coefficients do not depend on \(t\) and under certain assumption also in terms of coefficients. In particular, the result shows that the increments of the solution are ``asymptotically stationary. In Chapter 9 a result concerning stochastic flows due to Kunita is formulated. Finally, two versions of a multiparameter case are briefly exposed in Chapter 10, one due to \textit{X. Guyon} and \textit{B. Prum} [Z. Wahrscheinlichkeitstheorie Verw. Geb. 56, 361-397 (1981; Zbl 0461.60066)] and \textit{J. Yeh} [in: Real and stochastic analysis, 249-344 (1986; Zbl 0601.60055)] and another one due to \textit{J. B. Walsh} [in: École d'été de probabilités de Saint-Flour XIV, 1984. Lect. Notes Math. 1180, 265-437 (1986; Zbl 0608.60060)]. Generally, it is shown that many properties of ``usual SDEs actually remain valid for more general equations than Itô ones. Since the main and only example presented in the paper is still a Brownian motion case we would like to give rather briefly a few additional references close to the subject. Hypoellipticity results due to \textit{L. Hörmander} [Acta Math. 119 (1967), 147-171 (1968; Zbl 0156.10701) and ``The analysis of partial differential operators. III: Pseudo-differential operators (1985; Zbl 0601.35001)] could be mentioned as a classical PDE theory which can be used in such an analysis for degenerate SDEs, as well as its stochastic analogue, the Malliavin calculus. Weak solutions for partially degenerated SDEs (for which a 2nd or \(n\)th order one is a particular case) were studied by \textit{M. Nisio} [Osaka J. Math. 10, 185-208 (1973; Zbl 0268.60057)]. Strong solutions for a similar case were established by the reviewer [Izv. Akad. Nauk SSSR, Ser. Mat. 47, No. 1, 189-196 (1983; Zbl 0514.60056)]. The paper in a whole is an expanded survey which also poses certain very interesting questions, for example, concerning ergodic properties of the second order SDEs. For the entire collection see [Zbl 0947.00507].
- Generalized stochastic processes. Modelling and application of technical noise processes
- Generalized stochastic processes. Modelling and applications of noise processes. Translated from the German
- scientific article; zbMATH DE number 861520
- Stochastic processes. Translated from the Russian
- scientific article; zbMATH DE number 1738335
- Stochastic differential equations. An introduction with applications
- Theory and applications of stochastic processes. An analytical approach
- scientific article; zbMATH DE number 3888631
- Stochastic differential equations in science and engineering. With CD-ROM.
- Stochastic processes
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 Q4389736)
