Regularization for Continuously Observed Ordinal Response Variables with Piecewise-Constant Functional Predictors
Abstract
Many engineering applications involve the collection of functional data in which the unit of measurement is a curve measured over a continuous time domain. In this manuscript, we consider a case study involving a Follow-On Operational Test and Evaluation (FOT and E) conducted by the United States Army on the RQ-7BV2 Shadow, a tactical unmanned air vehicle system tasked with, among other tasks, providing full motion video to supported ground units. The quality of full motion video was the response variable of interest, and it was recorded on an ordinal scale continuously over the duration of each mission.1 In addition, important covariates, such as the air vehicles altitude and the distance from the air vehicle to the supported unit, were also observed continuously over time.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2016
- Accession Number
- AD1002101
Entities
People
- Laura Freeman
- Mark Orndorff
- Matthew Avery
- Timothy J Robinson
Organizations
- Institute for Defense Analyses