Refik Soyer - Profile on Academia.edu (original) (raw)
Papers by Refik Soyer
Decision Analysis, 2021
We provide an adversarial risk analysis framework for batch acceptance problems in which a decisi... more We provide an adversarial risk analysis framework for batch acceptance problems in which a decision maker relies exclusively on the size of the batch to accept or reject its admission to a system, albeit being aware of the presence of an opponent. The adversary acts as a data-fiddler attacker perturbing the observations perceived by the decision maker through injecting faulty items and/or modifying the existing items to faulty ones. We develop optimal policies against this combined attack strategy and illustrate the methodology with a review spam example.
Modeling Count Time Series
Chapman and Hall/CRC eBooks, Jul 12, 2022
Details of R-INLA for Time Series
Chapman and Hall/CRC eBooks, Jul 12, 2022
The American Statistician, 2019
We consider the fundamental problem of hypothesis testing extended by including the decisions of ... more We consider the fundamental problem of hypothesis testing extended by including the decisions of an adversary which aims at distorting the relevant data process observed so as to confound the decision maker, thus attaining a certain benefit. We provide an adversarial risk analysis approach to this problem and illustrate its usage in a batch acceptance context.
Bayesian Computations for Reliability Analysis in Dynamic Environments
Springer eBooks, Dec 9, 2021
The American Statistician, Jul 10, 2019
We consider the fundamental problem of hypothesis testing extended by including the decisions of ... more We consider the fundamental problem of hypothesis testing extended by including the decisions of an adversary which aims at distorting the relevant data process observed so as to confound the decision maker, thus attaining a certain benefit. We provide an adversarial risk analysis approach to this problem and illustrate its usage in a batch acceptance context.
Decision Analysis, Mar 1, 2021
We provide an adversarial risk analysis framework for batch acceptance problems in which a decisi... more We provide an adversarial risk analysis framework for batch acceptance problems in which a decision-maker relies exclusively on the size of the batch to accept or reject its admission to a system, while being aware of the presence of an opponent. The adversary acts as a data-fiddler attacker perturbing the observations perceived by the decision-maker through injecting faulty items and/or modifying the existing items to faulty ones. We develop optimal policies against this combined attack strategy and illustrate the methodology with a review spam example.
Ensuring Software Reliability
Technometrics, Feb 1, 1995
Discussion of “Virtual age, is it real?”
Applied Stochastic Models in Business and Industry, 2020
Discussion of “Virtual age: Is it real?”
Applied Stochastic Models in Business and Industry, 2021
Journal of the Operational Research Society, 2020
In this paper, we propose a Bayesian approach to model uncertainty in the bid arrival time by foc... more In this paper, we propose a Bayesian approach to model uncertainty in the bid arrival time by focusing on the time of the first bid in secondary (retail) market online business-to-business auctions. The proposed model is based on a Bayesian finite mixture of beta distributions. Our main objectives is to study potential heterogeneity of different auctions. In doing so, we incorporate some auction-specific features into the model and analyze their effect on the first bid time. We consider multiple competing models both in terms of fit and predictive performance. We also discuss managerial implications of the study and suggest how auctioneers can benefit from both the explanatory and predictive aspects of the model.
Information Importance of Models and Relative Importance of Predictors: Concept, Measures, Bayes Inference, and Applications
Comparison of relative importance of predictors is a subject of discussion of research findings i... more Comparison of relative importance of predictors is a subject of discussion of research findings in many disci- plines, as well as being input for decision-making in business practice. Relative importance methodologists have proposed measures for specific problems such as normal linear regression and logit. Some attempts have been made to set requirements for relative importance of predictors, given a measure of "importance", without characterizing the notion of "importance" itself. The main objective of this paper is to fill this gap by providing a notion of importance of predictors suciently general so as to be applicable to various models and data types, yet to admit a unique interpretation. The importance of predictors is characterized by the ex- tent to which their use reduces uncertainty about predicting the response variable, namely their information importance. Uncertainty associated with a probability distribution is a concave function of the density such...
Bayesian Modeling of Time Series of Counts with Business Applications
A semiparametric Bayesian model for queueing arrival processes: An application to call centers
Production and Operations Management
Nonhomogeneous Poisson process models have commonly been used to analyze and forecast arrivals. S... more Nonhomogeneous Poisson process models have commonly been used to analyze and forecast arrivals. Such processes require specification of intensity (arrival rate) functions, which are typically defined in a parametric form. The accuracy of the parametric models is highly sensitive to the choice of the specific intensity function for the arrival process. We use a Bayesian framework by proposing a nonparametric form for the intensity function and introduce a robust semiparametric model. The model is suitable for analyzing both time of arrival data and interval censored count data and can capture both monotonic and non‐monotonic arrival intensity. The intensity function in the model can be modulated to incorporate auxiliary information as well as seasonal and random effect components. We develop the Bayesian analysis of the proposed model and implement it on two real call center datasets with different characteristics. We also consider several extensions to our model and develop their Ba...
A latent‐factor self‐exciting point process for software failures
Naval Research Logistics (NRL)
Software debugging is the process of detecting and removing bugs during software development. Alt... more Software debugging is the process of detecting and removing bugs during software development. Although the intent of modifications to the software is to remove bugs, one cannot rule out the possibility of introducing new bugs as a result of these modifications. We consider a self‐exciting point process, which can incorporate the case of reliability deterioration due to the potential introduction of new bugs to the software during the development phase. In order to account for the unobservable process of introducing bugs, latent variables are incorporated into the self‐exciting point process models. The models are then applied to two data sets in software reliability and additional insights that can be obtained from these models are discussed. Our results suggest that the self‐exciting processes with latent factors perform better than the standard point process models in describing the behavior of software failures during the debugging process.
Bayesian Analysis of Proportions via a Hidden Markov Model
Methodology and Computing in Applied Probability
Dynamic Time Series Models using R-INLA
Bayesian Analysis of Doubly Stochastic Markov Processes in Reliability
Probability in the Engineering and Informational Sciences, 2020
Markov processes play an important role in reliability analysis and particularly in modeling the ... more Markov processes play an important role in reliability analysis and particularly in modeling the stochastic evolution of survival/failure behavior of systems. The probability law of Markov processes is described by its generator or the transition rate matrix. In this paper, we suppose that the process is doubly stochastic in the sense that the generator is also stochastic. In our model, we suppose that the entries in the generator change with respect to the changing states of yet another Markov process. This process represents the random environment that the stochastic model operates in. In fact, we have a Markov modulated Markov process which can be modeled as a bivariate Markov process that can be analyzed probabilistically using Markovian analysis. In this setting, however, we are interested in Bayesian inference on model parameters. We present a computationally tractable approach using Gibbs sampling and demonstrate it by numerical illustrations. We also discuss cases that invol...
Bayesian modeling of multivariate time series of counts
WIREs Computational Statistics, 2021
In this article, we present an overview of recent advances in Bayesian modeling and analysis of m... more In this article, we present an overview of recent advances in Bayesian modeling and analysis of multivariate time series of counts. We discuss basic modeling strategies including integer valued autoregressive processes, multivariate Poisson time series and dynamic latent factor models. In so doing, we make a connection with univariate modeling frameworks such as dynamic generalized models, Poisson state space models with gamma evolution and present Bayesian approaches that extend these frameworks to multivariate setting. During our development, recent Bayesian approaches to the analysis of integer valued autoregressive processes and multivariate Poisson models are highlighted and concepts such as “decouple/recouple” and “common random environment” are presented. The role that these concepts play in Bayesian modeling and analysis of multivariate time series are discussed. Computational issues associated with Bayesian inference and forecasting from these models are also considered.Thi...
Proceedings of the International Symposium on the Analytic Hierarchy Process, 1994
In this paper we present preliminary research on how information (or entropy) based measures can ... more In this paper we present preliminary research on how information (or entropy) based measures can be used by a decision maker (DM) using the Analytic Hierarchy Process (AHP) to assess judgment accuracy, and to decide when to stop the process of pairwise comparisons. We introduce some information indices, and illustrate their use in the Analytic Hierarchy Process by means of some examples. Furthermore, we research the issue of how to measure the information content of redundant judgments, and investigate the relationship between information and degree of redundancy through the use of simulated judgments.
Decision Analysis, 2021
We provide an adversarial risk analysis framework for batch acceptance problems in which a decisi... more We provide an adversarial risk analysis framework for batch acceptance problems in which a decision maker relies exclusively on the size of the batch to accept or reject its admission to a system, albeit being aware of the presence of an opponent. The adversary acts as a data-fiddler attacker perturbing the observations perceived by the decision maker through injecting faulty items and/or modifying the existing items to faulty ones. We develop optimal policies against this combined attack strategy and illustrate the methodology with a review spam example.
Modeling Count Time Series
Chapman and Hall/CRC eBooks, Jul 12, 2022
Details of R-INLA for Time Series
Chapman and Hall/CRC eBooks, Jul 12, 2022
The American Statistician, 2019
We consider the fundamental problem of hypothesis testing extended by including the decisions of ... more We consider the fundamental problem of hypothesis testing extended by including the decisions of an adversary which aims at distorting the relevant data process observed so as to confound the decision maker, thus attaining a certain benefit. We provide an adversarial risk analysis approach to this problem and illustrate its usage in a batch acceptance context.
Bayesian Computations for Reliability Analysis in Dynamic Environments
Springer eBooks, Dec 9, 2021
The American Statistician, Jul 10, 2019
We consider the fundamental problem of hypothesis testing extended by including the decisions of ... more We consider the fundamental problem of hypothesis testing extended by including the decisions of an adversary which aims at distorting the relevant data process observed so as to confound the decision maker, thus attaining a certain benefit. We provide an adversarial risk analysis approach to this problem and illustrate its usage in a batch acceptance context.
Decision Analysis, Mar 1, 2021
We provide an adversarial risk analysis framework for batch acceptance problems in which a decisi... more We provide an adversarial risk analysis framework for batch acceptance problems in which a decision-maker relies exclusively on the size of the batch to accept or reject its admission to a system, while being aware of the presence of an opponent. The adversary acts as a data-fiddler attacker perturbing the observations perceived by the decision-maker through injecting faulty items and/or modifying the existing items to faulty ones. We develop optimal policies against this combined attack strategy and illustrate the methodology with a review spam example.
Ensuring Software Reliability
Technometrics, Feb 1, 1995
Discussion of “Virtual age, is it real?”
Applied Stochastic Models in Business and Industry, 2020
Discussion of “Virtual age: Is it real?”
Applied Stochastic Models in Business and Industry, 2021
Journal of the Operational Research Society, 2020
In this paper, we propose a Bayesian approach to model uncertainty in the bid arrival time by foc... more In this paper, we propose a Bayesian approach to model uncertainty in the bid arrival time by focusing on the time of the first bid in secondary (retail) market online business-to-business auctions. The proposed model is based on a Bayesian finite mixture of beta distributions. Our main objectives is to study potential heterogeneity of different auctions. In doing so, we incorporate some auction-specific features into the model and analyze their effect on the first bid time. We consider multiple competing models both in terms of fit and predictive performance. We also discuss managerial implications of the study and suggest how auctioneers can benefit from both the explanatory and predictive aspects of the model.
Information Importance of Models and Relative Importance of Predictors: Concept, Measures, Bayes Inference, and Applications
Comparison of relative importance of predictors is a subject of discussion of research findings i... more Comparison of relative importance of predictors is a subject of discussion of research findings in many disci- plines, as well as being input for decision-making in business practice. Relative importance methodologists have proposed measures for specific problems such as normal linear regression and logit. Some attempts have been made to set requirements for relative importance of predictors, given a measure of "importance", without characterizing the notion of "importance" itself. The main objective of this paper is to fill this gap by providing a notion of importance of predictors suciently general so as to be applicable to various models and data types, yet to admit a unique interpretation. The importance of predictors is characterized by the ex- tent to which their use reduces uncertainty about predicting the response variable, namely their information importance. Uncertainty associated with a probability distribution is a concave function of the density such...
Bayesian Modeling of Time Series of Counts with Business Applications
A semiparametric Bayesian model for queueing arrival processes: An application to call centers
Production and Operations Management
Nonhomogeneous Poisson process models have commonly been used to analyze and forecast arrivals. S... more Nonhomogeneous Poisson process models have commonly been used to analyze and forecast arrivals. Such processes require specification of intensity (arrival rate) functions, which are typically defined in a parametric form. The accuracy of the parametric models is highly sensitive to the choice of the specific intensity function for the arrival process. We use a Bayesian framework by proposing a nonparametric form for the intensity function and introduce a robust semiparametric model. The model is suitable for analyzing both time of arrival data and interval censored count data and can capture both monotonic and non‐monotonic arrival intensity. The intensity function in the model can be modulated to incorporate auxiliary information as well as seasonal and random effect components. We develop the Bayesian analysis of the proposed model and implement it on two real call center datasets with different characteristics. We also consider several extensions to our model and develop their Ba...
A latent‐factor self‐exciting point process for software failures
Naval Research Logistics (NRL)
Software debugging is the process of detecting and removing bugs during software development. Alt... more Software debugging is the process of detecting and removing bugs during software development. Although the intent of modifications to the software is to remove bugs, one cannot rule out the possibility of introducing new bugs as a result of these modifications. We consider a self‐exciting point process, which can incorporate the case of reliability deterioration due to the potential introduction of new bugs to the software during the development phase. In order to account for the unobservable process of introducing bugs, latent variables are incorporated into the self‐exciting point process models. The models are then applied to two data sets in software reliability and additional insights that can be obtained from these models are discussed. Our results suggest that the self‐exciting processes with latent factors perform better than the standard point process models in describing the behavior of software failures during the debugging process.
Bayesian Analysis of Proportions via a Hidden Markov Model
Methodology and Computing in Applied Probability
Dynamic Time Series Models using R-INLA
Bayesian Analysis of Doubly Stochastic Markov Processes in Reliability
Probability in the Engineering and Informational Sciences, 2020
Markov processes play an important role in reliability analysis and particularly in modeling the ... more Markov processes play an important role in reliability analysis and particularly in modeling the stochastic evolution of survival/failure behavior of systems. The probability law of Markov processes is described by its generator or the transition rate matrix. In this paper, we suppose that the process is doubly stochastic in the sense that the generator is also stochastic. In our model, we suppose that the entries in the generator change with respect to the changing states of yet another Markov process. This process represents the random environment that the stochastic model operates in. In fact, we have a Markov modulated Markov process which can be modeled as a bivariate Markov process that can be analyzed probabilistically using Markovian analysis. In this setting, however, we are interested in Bayesian inference on model parameters. We present a computationally tractable approach using Gibbs sampling and demonstrate it by numerical illustrations. We also discuss cases that invol...
Bayesian modeling of multivariate time series of counts
WIREs Computational Statistics, 2021
In this article, we present an overview of recent advances in Bayesian modeling and analysis of m... more In this article, we present an overview of recent advances in Bayesian modeling and analysis of multivariate time series of counts. We discuss basic modeling strategies including integer valued autoregressive processes, multivariate Poisson time series and dynamic latent factor models. In so doing, we make a connection with univariate modeling frameworks such as dynamic generalized models, Poisson state space models with gamma evolution and present Bayesian approaches that extend these frameworks to multivariate setting. During our development, recent Bayesian approaches to the analysis of integer valued autoregressive processes and multivariate Poisson models are highlighted and concepts such as “decouple/recouple” and “common random environment” are presented. The role that these concepts play in Bayesian modeling and analysis of multivariate time series are discussed. Computational issues associated with Bayesian inference and forecasting from these models are also considered.Thi...
Proceedings of the International Symposium on the Analytic Hierarchy Process, 1994
In this paper we present preliminary research on how information (or entropy) based measures can ... more In this paper we present preliminary research on how information (or entropy) based measures can be used by a decision maker (DM) using the Analytic Hierarchy Process (AHP) to assess judgment accuracy, and to decide when to stop the process of pairwise comparisons. We introduce some information indices, and illustrate their use in the Analytic Hierarchy Process by means of some examples. Furthermore, we research the issue of how to measure the information content of redundant judgments, and investigate the relationship between information and degree of redundancy through the use of simulated judgments.