Use this URL to cite or link to this record in EThOS:
Title: Dynamic modelling of optimal pricing and trading policies under uncertainty
Author: Abbaszadeh, Shahin
ISNI:       0000 0004 5370 7472
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2015
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
The objective of this thesis is to present a set of useful tools for problems of sequential decision making under uncertainty. Specifically, we study three applications of dynamic planning: dynamic pricing of non-durable products in the context of Markov processes, dynamic pricing of high end fashionable products with autoregressive demand, and the dynamic trading of financial securities with added sign constraints. Market volatility, incomplete or delayed information, and unpredictability of underlying systems are integral to real-world problems. It is important to establish methods to integrate these factors into the modelling framework of choice. In this research we study stochastic dynamic programs and their use in finding optimal or near-optimal strategies for the above problems. In the first of three papers comprising this thesis, we examine the dynamic pricing problem in the context of Markov decision processes, and explore the structural characteristics of the model. Our results support the use of exact methods when assuming the state of the system (demand) is unobservable. The second paper is concerned with a dynamic pricing problem that assumes an autoregressive evolution model for the demand. We provide a simple but ef- fective approximate dynamic programming method that outperforms the classic methods of solving dynamic programming problems. Finally, in the third paper, we examine the dynamic trading of large blocks of securities by extending the dynamic programming framework to include constraints and additional information. We explore the characteristics of the model to improve on the closed form solutions available in the literature, but we also utilise a heuristic approximate dynamic programming method to provide near-optimal results when the problem is augmented with necessary constraints to handle practical settings.
Supervisor: Wu, Yue Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: HF Commerce