Stochastic differential equations: singularity of coefficients, regression models, and stochastic approximation (Q4375215)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 1111340
Language Label Description Also known as
default for all languages
No label defined
    English
    Stochastic differential equations: singularity of coefficients, regression models, and stochastic approximation
    scientific article; zbMATH DE number 1111340

      Statements

      Stochastic differential equations: singularity of coefficients, regression models, and stochastic approximation (English)
      0 references
      1 February 1998
      0 references
      semimartingales
      0 references
      stochastic differential equations
      0 references
      stochastic approximation
      0 references
      law of large numbers
      0 references
      law of the iterated logarithm
      0 references
      asymptotic normality
      0 references
      0 references
      This survey article is devoted to the application of the stochastic calculus for semimartingales to statistical approximation and models of financial mathematics. Correspondingly to this aim the article is divided into four parts. The first part contains the limit theorems for the continuous time martingales. Here the strong law of the large numbers and the law of the iterated logarithm are presented. In the second chapter the stochastic differential equations driven by semimartingales are considered. The main object is the following equation \[ X_t=X_0+\int_0^tf(s,X_s) dA_s+\int_0^tg(s,X_s) dM_s.\tag{1} \] Here \(A\) is a predictable continuous increasing process, \(M\) is a local continuous martingale. The main result of this chapter is the existence of the strong solution for the equation (1) where the drift coefficient \(f\) does not satisfy the Lipschitz condition. For to get this statement the author proves the analog of the Skorokhod and Yamada comparison theorem for the equation of type (1) and then uses the method of monotone approximation. In the third chapter the generalized (continuous time) Robbins-Monro and Kiefer-Wolfowitz procedures are considered as solution of certain stochastic differential equations. Using the limit theorems for the martingales from the previous chapters the author proves the convergence of this procedures and their asymptotic normality. Chapter four contains some regression models and models from the finance which lead to stochastic differential equations driven by martingales and statements about the limit behavior of the corresponding estimates.
      0 references

      Identifiers