Talias, Michael Andrea; (2000) Statistical and financial aspects of research and development portfolios. Doctoral thesis (Ph.D), UCL (University College London).

Text Statistical_and_financial_aspe.pdf Download (3MB)

Abstract

Sequential evaluation and decision problems must frequently be solved under uncertainty. The sequential nature of activities like research, development or exploration, requires optimal funding criteria that take into account the fact that further funding decisions will be made throughout the future. In this thesis, we examine several sequential and parallel strategies for R and D project selection and capital budgeting problems. Some of these problems have as a solution a prioritisation index. We pay particular interest to the Pearson and Gittins indices. We relate the Pearson index to the Neyman-Pearson lemma and state clearly the kind of Problems the Pearson index solves. We reformulate this problem using non-linear utility function and show how to solve it for different utility functions. These kind of indices may need to have a forecast for R and D rewards or costs. We discuss adaptive prediction, we derive the forecasting rule for various data generating processes, and study the behaviour of unconditional and conditional forecast variances. Furthermore, we study the connection of R and D projects with real options theory, and discuss the suitability of this methodology and its fundamental principle of economic rationality or no-arbitrage. Finally, the multi-armed bandit problem is introduced and is reconciled with the option pricing. We prove that an additional condition is required for an index policy to be optimal when two projects are selected simultaneously with criterion the sum of their indices to be maximum.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as JSONCSVXML

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item