- Library Home /
- Search Collections /
- Open Collections /
- Browse Collections /
- BIRS Workshop Lecture Videos /
- Bootstrap Robust Prescriptive Analytics
Open Collections
BIRS Workshop Lecture Videos
BIRS Workshop Lecture Videos
Bootstrap Robust Prescriptive Analytics Van Parys, Bart Paul Gerard
Description
We discuss prescribing optimal decisions in a framework where their cost depends on uncertain problem parameters that need to be learned from supervised data. Prescriptive analytics consists in making optimal decisions specific to a particular covariate context. Prescriptive methods need to factor in additional observed contextual information on a potentially large number of covariates as opposed to static decision methods who only use sample data. Any naive use of training data may, however, lead to gullible decisions over-calibrated to one particular data set. In this presentation, we use ideas from distributionally robust optimization and the statistical bootstrap to propose two novel prescriptive methods based on (nw) Nadaraya-Watson and (nn) nearest neighbors learning which safeguard against overfitting and lead to improved out-of-sample performance. Both resulting robust prescriptive methods reduce to tractable convex optimization problems and enjoy a limited disappointment on bootstrap data. We illustrate the data-driven decision-making framework and our novel robustness notion on a small news vendor problem as well as a small portfolio allocation problem.
Item Metadata
Title |
Bootstrap Robust Prescriptive Analytics
|
Creator | |
Publisher |
Banff International Research Station for Mathematical Innovation and Discovery
|
Date Issued |
2018-03-06T14:03
|
Description |
We discuss prescribing optimal decisions in a framework where their cost depends on uncertain problem parameters that need to be learned from supervised data. Prescriptive analytics consists in making optimal decisions specific to a particular covariate context. Prescriptive methods need to factor in additional observed contextual information on a potentially large number of covariates as opposed to static decision methods who only use sample data. Any naive use of training data may, however, lead to gullible decisions over-calibrated to one particular data set. In this presentation, we use ideas from distributionally robust optimization and the statistical bootstrap to propose two novel prescriptive methods based on (nw) Nadaraya-Watson and (nn) nearest neighbors learning which safeguard against overfitting and lead to improved out-of-sample performance. Both resulting robust prescriptive methods reduce to tractable convex optimization problems and enjoy a limited disappointment on bootstrap data. We illustrate the data-driven decision-making framework and our novel robustness notion on a small news vendor problem as well as a small portfolio allocation problem.
|
Extent |
33 minutes
|
Subject | |
Type | |
File Format |
video/mp4
|
Language |
eng
|
Notes |
Author affiliation: Massachusetts Institute of Technology
|
Series | |
Date Available |
2018-09-03
|
Provider |
Vancouver : University of British Columbia Library
|
Rights |
Attribution-NonCommercial-NoDerivatives 4.0 International
|
DOI |
10.14288/1.0371890
|
URI | |
Affiliation | |
Peer Review Status |
Unreviewed
|
Scholarly Level |
Postdoctoral
|
Rights URI | |
Aggregated Source Repository |
DSpace
|
Item Media
Item Citations and Data
Rights
Attribution-NonCommercial-NoDerivatives 4.0 International