Abstract
In this chapter we deal with the dynamic programming algorithm DPA for the finite-horizon stochastic dynamic programming problem SDP introduced in Section 4.6. Roughly speaking, the following decision process is described by SDP. There is a discrete time stochastic system, and the state of the system evoluates in a Markovian way. At each stage of time the current state is observed, and using this information the decision maker has to select an action. Any action will influence the immediate cost of the corresponding stage as well as the probability distribution of the next state. The purpose is to minimize the expected costs, summed over all stages. For details on the SDP problem we refer the reader to Section 4.6. There we gave also a formal definition of a “policy”; a policy may be seen as a complete specification of all particular choices which possibly are made by the decision maker at any stage and at any state.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
K.J. Arrow, S. Karlin, H. Scare (1958). Studies in the Mathematical Theory of Inventory and Production, Stanford University Press, Stanford Ca.
D.P. Bertsekas, S.E. Shreve (1978). Stochastic Optimal Control; the Discrete Time Case, Academic Press, New York.
D. Blackwell, D. Freedman, M. Orkin (1974). The optimal reward operator in dynamic programming. Ann. Probab. 2, 926–941.
I.V. Evstigneev (1976). Measurable selection and dynamic programming. Math. Oper. Res. 1, 267–272.
D.A. Freedman (1974). The optimal reward operator in special cases of dynamic programming problems. Ann. Prob. 2, 942–949.
K. Hinderer (1970). Foundations of Non-stationary Dynamic Programming with Discrete Time Parameter. Lecture notes in operations research and mathematical systems 33, Springer, Berlin-Heidelberg-New York.
W.K. Klein Haneveld (1977). Markovian Inventory Control Models, Report Institute of Econometrics OR 7702, University of Groningen.
W.K. Klein Haneveld (1978). Markovian Inventory Control Models; Improvements and Extensions. Report Department of Mathematics, University of Kentucky, Lexington Ky.
P. Olsen (1976). When is a multistage stochastic programming problem well-defined? SIAM J. Control Optim. 14, 518–527.
R.T. Rockafellar (1976). Integral functionals, normal integrands and measurable selection. J.P. Gossez, E.J. Lami Dozo, J. Mawhin, L. Waelbroeck (eds.). Nonlinear Operators and the Calculus of Variation, Lecture notes in mathematics 543, Springer, Berlin-Heidelberg-New York, 157–207.
R.T. Rockafellar, R.J.-B. Wets (1976). Nonanticipativity and L1-martingales in stochastic optimization problems. Math. Programming Stud. 6, 170–187.
M. Schäl (1972). On continuous dynamic programming with discrete time parameter. Z. Wahrsch. Verw. Gebiete 21, 279–288.
M. Schäl (1975). On dynamic programming: compactness of the space of policies. Stochastic Process. Appl. 3, 345–364.
M. Schäl (1976). On the optimality of (s,S)-policies In Dynamic Inventory models with finite horizon. SIAM J. Appl. Math. 30, 528–537.
S. Shreve (1977). Dynamic Programming in Complete Separable Spaces. Ph.D. thesis, University of Illinois at Urbana-Champaign, Urbana Ill.
R.E. Strauch (1966). Negative dynamic programming. Ann. Statist. 37, 871–890.
R.J.-B. Wets (1976). Grundlagen Konvexer Optimierung. Lecture notes in economics and mathematical systems 137, Springer, Berlin-Heidelberg-New York.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1986 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Klein Haneveld, W.K. (1986). On the Behaviour of the Optimal Value Operator of Dynamic Programming. In: Duality in Stochastic Linear and Dynamic Programming. Lecture Notes in Economics and Mathematical Systems, vol 274. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-51697-9_6
Download citation
DOI: https://doi.org/10.1007/978-3-642-51697-9_6
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-16793-8
Online ISBN: 978-3-642-51697-9
eBook Packages: Springer Book Archive