Health and loan default risk analytics : prediction models and their evaluation
Abstract/Contents
- Abstract
- Risk prediction of future events or other adverse outcomes is widely used in health and medicine, economics and finance, climate change and infrastructure reliability. How to evaluate the risk forecasts and forecasting models/methods and quantify the difference of the predictions from the true risks is an important but challenging problem in using the forecasts to guide regulatory, or policy, or treatment/intervention decisions. Lai et al (2011) review the literature on the evaluation of risk forecasts using scoring rules and associated loss functions in weather forecasting and econometric forecasts, and introduce a new approach that gives valid confidence intervals for the difference in average scores of two forecasts. It provides a martingale approach to analyzing the prediction accuracy for binary outcomes when the outcomes are not independent across time and space. In this thesis, we give some background of score statistics and concordance-type statistics, discuss their limitations and some challenges in the evaluation of risk models, and develop methods, more specifically, the martingale approach, to address these challenges. We also extend our approach to a) subgroup identification using multiple hypotheses testing framework and b) sequential test to find change point of model performance statistics. We demonstrate our methods with simulations and two real-world applications.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2019; ©2019 |
| Publication date | 2019; 2019 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Miao, Jing |
|---|---|
| Degree supervisor | Lai, T. L |
| Thesis advisor | Lai, T. L |
| Thesis advisor | Lu, Ying, 1960- |
| Thesis advisor | Walther, Guenther |
| Degree committee member | Lu, Ying, 1960- |
| Degree committee member | Walther, Guenther |
| Associated with | Stanford University, Department of Statistics. |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Jing Miao. |
|---|---|
| Note | Submitted to the Department of Statistics. |
| Thesis | Thesis Ph.D. Stanford University 2019. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2019 by Jing Miao
- 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...