The rate function for some measure-valued jump processes (Q1902958)

The purpose of this paper is to derive the large deviation principle in the McKean-Vlasov limit and to obtain several representations of the associated large deviation rate function. The finite systems are of the form \(x = (x^1,\dots, x^n) \in E^n\), where to each component corresponds a jump process \(t \mapsto x^i_t\), with values in the state space \(E \subset \mathbb{R}^d\). A typical example is that \(E = \mathbb{Z}_+\), the nonnegative integers, and the \(x^i\) represent the number of particles present at each of \(n\) reaction sites. Schlögl type models with meanfield interaction have been studied in depth. \textit{D. A. Dawson} and \textit{X. Zheng} [Adv. Appl. Math. 12, No. 3, 293-326 (1991; Zbl 0751.60080)] and \textit{S. Feng} and \textit{X. Zheng} [Stochastic Processes Appl. 43, No. 1, 65-84 (1992; Zbl 0762.60075)] obtained existence results and studied various properties of the McKean-Vlasov limit. \textit{S. Feng} [Ann. Probab. 22, No. 4, 2122-2151 (1994)] considered the large deviation problem. As for applications, epidemic systems with emphasis on limit behavior and large deviations were studied by \textit{C. Léonard} [C. R. Acad. Sci., Paris, Sér. I 308, No. 13, 425-428 (1989; Zbl 0661.60044) and Lect. Notes Biomath. 86]. The main references are the works [\textit{D. A. Dawson} and \textit{J. Gärtner}, Stochastics 20, 247-308 (1987; Zbl 0613.60021); \textit{J. Gärtner}, Math. Nachr. 137, 197-248 (1988; Zbl 0678.60100)] on weakly interacting diffusions. A motivation of this paper was to unify some results obtained by Feng (loc. cit.) and Léonard (loc. cit.) and possibly extend them in the direction to which the results of Dawson and Gärtner (loc. cit.) point.