1: \begin{abstract}  

2: This paper studies a structured compound stochastic program (SP) involving multiple

3: expectations coupled by nonconvex and nonsmooth functions.  

4: We present a successive convex-programming based sampling algorithm and establish its subsequential convergence.

5: We describe stationarity properties of the limit points for several classes of the compound SP.  We further 

6: discuss probabilistic stopping rules based on the computable error-bound for the algorithm.

7: % We present a successive convex-programming based sampling algorithm for solving this problem, establish its

8: % subsequential convergence, and discuss probabilistic  stopping rules based on the computable

9: % error-bound for the algorithm.  

10: We present several risk measure minimization problems that can be formulated as such a compound stochastic program; 

11: these include generalized deviation optimization problems based on the optimized certainty

12: equivalent and buffered probability of exceedance (bPOE), a

13: distributionally robust bPOE optimization problem, and a

14: multiclass classification problem employing the cost sensitive error criteria with bPOE risk measure.\end{abstract}

15: