journal article
LitStream Collection
Issue Information
2018 Applied Stochastic Models in Business and Industry
doi: 10.1002/asmb.2279
No abstract is available for this article.
Home
2018 Applied Stochastic Models in Business and Industry
doi: 10.1002/asmb.2279
No abstract is available for this article.
Grelaud, Aude; Mitra, Priyam; Xie, Minge; Chen, Rong
2018 Applied Stochastic Models in Business and Industry
doi: 10.1002/asmb.2293
This paper proposes a dynamic system, with an associated fusion learning inference procedure, to perform real‐time detection and localization of nuclear sources using a network of mobile sensors. This is motivated by the need for a reliable detection system in order to prevent nuclear attacks in major cities such as New York City. The approach advocated here installs a large number of relatively inexpensive (and perhaps relatively less accurate) nuclear source detection sensors and GPS devices in taxis and police vehicles moving in the city. Sensor readings and GPS information are sent to a control center at a high frequency, where the information is immediately processed and fused with the earlier signals. We develop a real‐time detection and localization method aimed at detecting the presence of a nuclear source and estimating its location and power. We adopt a Bayesian framework to perform the fusion learning and use a sequential Monte Carlo algorithm to estimate the parameters of the model and to perform real‐time localization. A simulation study is provided to assess the performance of the method for both stationary and moving sources. The results provide guidance and recommendations for an actual implementation of such a surveillance system. Copyright © 2017 John Wiley & Sons, Ltd.
Myhre, Janet; Jeske, Daniel R.; Li, Jun; Hansen, Anne M.
2018 Applied Stochastic Models in Business and Industry
doi: 10.1002/asmb.2255
A commonly occurring problem in reliability testing is how to combine pass/fail test data that is collected from disparate environments. We have worked with colleagues in aerospace engineering for a number of years where two types of test environments in use are ground tests and flight tests. Ground tests are less expensive and consequently more numerous. Flight tests are much less frequent, but directly reflect the actual usage environment. We discuss a relatively simple combining approach that realizes the benefit of a larger sample size by using ground test data, but at the same time accounts for the difference between the two environments. We compare our solution with what look like more sophisticated approaches to the problem in order to calibrate its limitations. Overall, we find that our proposed solution is robust to its inherent assumptions, which explains its usefulness in practice. Copyright © 2017 John Wiley & Sons, Ltd.
Hoegh, Andrew; Leman, Scotland
2018 Applied Stochastic Models in Business and Industry
doi: 10.1002/asmb.2261
Model fusion methods, or more generally ensemble methods, are a useful tool for prediction. Combining predictions from a set of models smooths out biases and reduces variances of predictions from individual models, and hence, the combined predictions typically outperform those from individual models. In many algorithms, individual predictions are arithmetically averaged with equal weights. However, in the presence of correlated models, the fusion process is required to account for association between models; otherwise, the naively averaged predictions will be suboptimal. This article describes optimal model fusion principles and illustrates the potential pitfalls of naive fusion in the presence of correlated models for binary data. An efficient algorithm for correlated model fusion is detailed and applied to algorithms mining social media information to predict civil unrest. Copyright © 2017 John Wiley & Sons, Ltd.
Casleton, Emily; Osthus, Dave; Van Buren, Kendra
2018 Applied Stochastic Models in Business and Industry
doi: 10.1002/asmb.2299
Missing data are prevalent issue in analyses involving data collection. The problem of missing data is exacerbated for multisource analysis, where data from multiple sensors are combined to arrive at a single conclusion. In this scenario, it is more likely to occur and can lead to discarding a large amount of data collected; however, the information from observed sensors can be leveraged to estimate those values not observed. We propose two methods for imputation of multisource data, both of which take advantage of potential correlation between data from different sensors, through ridge regression and a state‐space model. These methods, as well as the common median imputation, are applied to data collected from a variety of sensors monitoring an experimental facility. Performance of imputation methods is compared with the mean absolute deviation; however, rather than using this metric to solely rank the methods, we also propose an approach to identify significant differences. Imputation techniques will also be assessed by their ability to produce appropriate confidence intervals, through coverage and length, around the imputed values. Finally, performance of imputed datasets is compared with a marginalized dataset through a weighted k‐means clustering. In general, we found that imputation through a dynamic linear model tended to be the most accurate and to produce the most precise confidence intervals, and that imputing the missing values and down weighting them with respect to observed values in the analysis led to the most accurate performance. Published 2017. This article is a U.S. Government work and is in the public domain in the USA.
Serhiyenko, Volodymyr; Ravishanker, Nalini; Venkatesan, Rajkumar
2018 Applied Stochastic Models in Business and Industry
doi: 10.1002/asmb.2232
This article describes statistical analyses pertaining to marketing data from a large multinational pharmaceutical firm. We describe models for monthly new prescription counts that are written by physicians for the firm's focal drug and for competing drugs, as functions of physician‐specific and time‐varying predictors. Modeling patterns in discrete‐valued time series, and specifically time series of counts, based on large datasets, is the focus of much recent research attention. We first provide a brief overview of Bayesian approaches we have employed for modeling multivariate count time series using Markov Chain Monte Carlo methods. We then discuss a flexible level correlated model framework, which enables us to combine different marginal count distributions and to build a hierarchical model for the vector time series of counts, while accounting for the association among the components of the response vector, as well as possible overdispersion. We employ the integrated nested Laplace approximation (INLA) for fast approximate Bayesian modeling using the R‐INLA package (r‐inla.org). To enhance computational speed, we first build a model for each physician, use features of the estimated trends in the time‐varying parameters in order to cluster the physicians into groups, and fit aggregate models for all physicians within each cluster. Our three‐stage analysis can provide useful guidance to the pharmaceutical firm on their marketing actions. Copyright © 2017 John Wiley & Sons, Ltd.
Showing 1 to 10 of 10 Articles