| Abstract: |
| This paper studies a general class of mean field games involving a major agent and a large number of minor agents, whose payoff functionals are \emph{recursive} and represented in terms of the solution of backward stochastic differential equations, referred to as recursive major-minor (RMM) problems. Our RMM modeling encompasses weak couplings of empirical averages into the recursive functionals and dynamics of both major and minor agents and incorporates
general non-additive functionals. The auxiliary limiting game of RMM is constructed via a novel mixed triple-agent leader-follower-Nash games. The associated consistency system is derived and related asymptotic major-minor equilibrium is constructed. In addition, linear-quadratic settings of RMM problems are studied to illustrate our results. |
|