Continuous approximation schemes for stochastic programs

One of the main methods for solving stochastic programs is approximation by discretizing the probability distribution. However, discretization may lose differentiability of expectational functionals. The complexity of discrete approximation schemes also increases exponentially as the dimension of the random vector increases. On the other hand, stochastic methods can solve stochastic programs with larger dimensions but their convergence is in the sense of probability one. In this paper, we study the differentiability property of stochastic two-stage programs and discuss continuous approximation methods for stochastic programs. We present several ways to calculate and estimate this derivative. We then design several continuous approximation schemes and study their convergence behavior and implementation. The methods include several types of truncation approximation, lower dimensional approximation and limited basis approximation.

[1]  Julia L. Higle,et al.  Stochastic Decomposition: An Algorithm for Two-Stage Linear Programs with Recourse , 1991, Math. Oper. Res..

[2]  Bennett L. Fox,et al.  Algorithm 647: Implementation and Relative Efficiency of Quasirandom Sequence Generators , 1986, TOMS.

[3]  András Prékopa,et al.  Boole-Bonferroni Inequalities and Linear Programming , 1988, Oper. Res..

[4]  Werner Römisch,et al.  Stability analysis for stochastic programs , 1991, Ann. Oper. Res..

[5]  István Deák,et al.  Three digit accurate multiple normal probabilities , 1980 .

[6]  Y. Ermoliev.  Stochastic quasigradient methods and their application to system optimization , 1983 .

[7]  J. Birge,et al.  Computing block-angular Karmarkar projections with applications to stochastic programming , 1988 .

[8]  R. Wets,et al.  Designing approximation schemes for stochastic optimization problems, in particular for stochastic programs with recourse , 1986 .

[9]  István Deák,et al.  Random Number Generators and Simulation , 1990 .

[10]  Roger J.-B. Wets,et al.  Stochastic programming 84 , 1986 .

[11]  Roger J.-B. Wets,et al.  Stochastic programming, an Introduction , 1988 .

[12]  Yuri Ermoliev,et al.  Numerical techniques for stochastic optimization , 1988 .

[13]  Roger J.-B. Wets,et al.  Sublinear upper bounds for stochastic programs with recourse , 1987, Math. Program..

[14]  Jinde Wang,et al.  Lipschitz continuity of objective functions in stochastic programs with fixed recourse and its applications , 1986 .

[15]  F. Clarke.  Optimization And Nonsmooth Analysis , 1983 .

[16]  C. Givens,et al.  A class of Wasserstein metrics for probability distributions. , 1984 .

[17]  Peter Kall,et al.  Stochastic Linear Programming , 1975 .

[18]  H. Gassmann.  Conditional Probability and Conditional Expectation of a Random Vector , 1988 .

[19]  S. Rachev.  The Monge–Kantorovich Mass Transference Problem and Its Stochastic Applications , 1985 .

[20]  Werner Römisch,et al.  Distribution sensitivity in stochastic programming , 1991, Math. Program..

[21]  John R. Birge,et al.  Decomposition and Partitioning Methods for Multistage Stochastic Linear Programs , 1985, Oper. Res..

[22]  Y. Smeers,et al.  Optimal Investments for Electricity Generation: A Stochastic Model and a Test-Problem , 1988 .

[23]  Roger J.-B. Wets,et al.  Stochastic Programming: Solution Techniques and Approximation Schemes , 1982, ISMP.

[24]  Roger J.-B. Wets,et al.  Computing Bounds for Stochastic Programming Problems by Means of a Generalized Moment Problem , 1987, Math. Oper. Res..

[25]  Liqun Qi,et al.  Semiregularity and Generalized Subdifferentials with Applications to Optimization , 1993, Math. Oper. Res..

[26]  Jinde Wang,et al.  Distribution sensitivity analysis for stochastic programs with complete recourse , 1985, Math. Program..

[27]  J. Birge,et al.  A separable piecewise linear upper bound for stochastic linear programs , 1988 .

[28]  Karl Frauendorfer,et al.  Stochastic Two-Stage Programming , 1992 .