Quantile function methods for decision analysis
Abstract/Contents
- Abstract
- Encoding prior probability distributions is a fundamental step in any decision analysis. A decision analyst often elicits an expert's knowledge about a continuous uncertain quantity as a set of quantile-probability pairs (points on a cumulative distribution function) and seeks a probability distribution consistent with them. Quantile-parameterized distributions (QPDs) are continuous probability distributions characterized by quantile-probability data. This dissertation demonstrates the flexibility of QPDs to represent a wide range of distributional shapes, examines a QPD's range of parametric feasibility, and introduces various means of using QPDs when encoding relevance between uncertainties. A decision maker may or may not believe a continuous uncertain quantity has well-defined bounds. For the former case, I offer a toolkit for engineering the support of a QPD. For the latter, I develop a theory for comparing tail heaviness between probability distributions and offer methods for engineering the tail behavior of a QPD. I conclude with an example decision analysis: a pharmaceutical company CEO's decision whether to market or license a drug. This analysis uses QPDs to encode prior probability distributions with bounded and unbounded supports. It introduces three tools that use QPDs: data compression of multivariate probabilistic simulation, sensitivity analysis of the tail heaviness of a prior probability distribution, and valuation of probability assessment.
Description
Type of resource | text |
---|---|
Form | electronic; electronic resource; remote |
Extent | 1 online resource. |
Publication date | 2013 |
Issuance | monographic |
Language | English |
Creators/Contributors
Associated with | Powley, Bradford W | |
---|---|---|
Associated with | Stanford University, Department of Management Science and Engineering. | |
Primary advisor | Howard, Ronald A. (Ronald Arthur), 1934- | |
Thesis advisor | Howard, Ronald A. (Ronald Arthur), 1934- | |
Thesis advisor | Keelin, Thomas W | |
Thesis advisor | Shachter, Ross D | |
Advisor | Keelin, Thomas W | |
Advisor | Shachter, Ross D |
Subjects
Genre | Theses |
---|
Bibliographic information
Statement of responsibility | Bradford W. Powley. |
---|---|
Note | Submitted to the Department of Management Science and Engineering. |
Thesis | Thesis (Ph.D.)--Stanford University, 2013. |
Location | electronic resource |
Access conditions
- Copyright
- © 2013 by Bradford William Powley
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...