Douglas C. Creighton

Learn More
This paper evaluates the four leading techniques proposed in the literature for construction of prediction intervals (PIs) for neural network point forecasts. The delta, Bayesian, bootstrap, and mean-variance estimation (MVE) methods are reviewed and their performance for generating high-quality PIs is compared. PI-based measures are proposed and applied(More)
Prediction intervals (PIs) have been proposed in the literature to provide more information by quantifying the level of uncertainty associated to the point forecasts. Traditional methods for construction of neural network (NN) based PIs suffer from restrictive assumptions about data distribution and massive computational loads. In this paper, we propose a(More)
The accurate prediction of travel times is desirable, but frequently prone to error. This is mainly attributable to both the underlying traffic processes, as well as in the data which are used to infer travel time. A more meaningful and pragmatic approach is to view travel time prediction as probabilistic inference and to construct prediction intervals,(More)
OLE Process Control (OPC) is an industry standard that facilitates the communication between PCs and Programmable Logic Controllers (PLC). This communication allows for the testing of control systems with an emulation model. When models require faster and higher volume communications, limitations within OPC prevent this. In this paper an interface is(More)
—Controlled mobility in wireless sensor networks provides many benefits towards enhancing the network performance and prolonging its lifetime. Mobile elements, acting as mechanical data carriers, traverse the network collecting data using single-hop communication, instead of the more energy demanding multi-hop routing to the sink. Scaling up from single to(More)
The bootstrap method is one of the most widely used methods in literature for construction of confidence and prediction intervals. This paper proposes a new method for improving the quality of bootstrap-based prediction intervals. The core of the proposed method is a prediction interval-based cost function, which is used for training neural networks. A(More)