Search results

This paper investigates the usefulness of switching Gaussian state space models as a tool for implementing dynamic model selection (DMS) or averaging (DMA) in time-varying parameter regression models. DMS methods allow for model switching, where a different model can be chosen at each point in time. Thus, they allow for the explanatory variables in the time-varying parameter regression model to change over time. DMA will carry out model averaging in a time-varying manner. We compare our exact method for implementing DMA/DMS to a popular existing procedure which relies on the use of forgetting factor approximations. In an application, we use DMS to select different predictors in an inflation forecasting application. We find strong evidence of model switching. We also compare different ways of implementing DMA/DMS and find forgetting factor approaches and approaches based on the switching Gaussian state space model to lead to similar results.

A challenging task for context‐aware pervasive systems is reasoning about context in uncertain environments where sensors can be inaccurate or unreliable and inferred situations ambiguous and uncertain. This paper aims to address this grand challenge, with research in context awareness to provide feasible solutions by means of theoretical models, algorithms and reasoning approaches.

This paper proposes a theoretical model about context and a set of context verification procedures, built over the model and implemented in a context reasoning engine prototype. The verification procedures utilize beneficial characteristics of spatial representation of context and also provide guidelines based on heuristics that lead to resolution of conflicts arising due to context uncertainty. The engine's reasoning process is presented and it is shown how the proposed modeling and verification approach contributes in tackling the uncertainty associated with the reasoning task. The paper experimentally evaluates this approach with a distributed simulation of a sensor‐based office environment with unreliable and inaccurate sensors.

Important features of the model are dynamic aspects of context, such as context trajectory and stability of a pervasive system in given context. These can also be used for context verification as well as for context prediction. The model strength is also in its generality and its ability to model a variety of context‐aware scenarios comprising different types of information.

The paper describes a theoretical model for context and shows it is useful not only for context representation but also for developing reasoning and verification techniques for uncertain context.

This paper is concerned with the state‐space approach to optimal control problems of dynamic econometric systems. We show how the state‐space approach can be integrated into the traditional econometric method, and how much could be gained by this consolidated approach.

The purpose of this paper is to examine the impact of applying two classical controller strategies, including two proportional (P) controllers with two feedback loops and one proportional–integral–derivative (PID) controller with one feedback loop, on the order and inventory performance within a production-inventory control system.

The simulation experiments of the dynamics behaviour of the production-inventory control system are conducted using a model based on control theory techniques. The Laplace transformation of an Order–Up–To (OUT) model is obtained using a state-space approach, and then the state-space representation is used to design and simulate a controlled model. The simulations of each model with two control configurations are tested by subjecting the system to a random retail sales pattern. The performance of inventory level is quantified by using the Integral of Absolute Error (IAE), whereas the bullwhip effect is measured by using the Variance ratio (Var).

The simulation results show that one PID controller with one feedback loop outperforms two P controllers with two feedback loops at reducing the bullwhip effect and regulating the inventory level.

The production-inventory control system is broken down into three components, namely: the forecasting mechanism, controller strategy and production-inventory process. A state-space approach is adopted to design and simulate the different controller strategy.

Active service is one of key problems of ubiquitous computing paradigm. Context‐aware computing is helpful to carry out this service. Because the context is changing with the movement or shift of the user, its uncertainty often exists. Context‐aware computing with uncertainty includes obtaining context information, forming model, fusing of aware context and managing context information. In this paper, we focus on modeling and computing of aware context information with uncertainty for making dynamic decision during seamless mobility. Our insight is to combine dynamic context‐aware computing with improved Random Set Theory (RST) and extended D‐S Evidence Theory (EDS). We re‐examine formalism of random set, argue the limitations of the direct numerical approaches, give new modeling mode based on RST for aware context and propose our computing approach of modeled aware context.In addition, we extend classic D‐S Evidence Theory after considering context’s reliability, time‐efficiency and relativity, compare relative computing methods. After enumerating experimental examples of our active space, we provide the evaluation. By comparisons, the validity of new context‐aware computing approach based on RST or EDS for ubiquitous active service with uncertainty information has been successfully tested.

We compare methods to measure comovement in business cycle data using multi-level dynamic factor models. To do so, we employ a Monte Carlo procedure to evaluate model performance for different specifications of factor models across three different estimation procedures. We consider three general factor model specifications used in applied work. The first is a single-factor model, the second a two-level factor model, and the third a three-level factor model. Our estimation procedures are the Bayesian approach of Otrok and Whiteman (1998), the Bayesian state-space approach of Kim and Nelson (1998) and a frequentist principal components approach. The latter serves as a benchmark to measure any potential gains from the more computationally intensive Bayesian procedures. We then apply the three methods to a novel new dataset on house prices in advanced and emerging markets from Cesa-Bianchi, Cespedes, and Rebucci (2015) and interpret the empirical results in light of the Monte Carlo results.

This paper offers various state‐space representations in the context of applications of the system control theory to dynamic economic systems and examines interrelationships between the alternative representations in both economics literature and system control engineering literature. In particular, some characteristics of various state‐space forms are assessed with respect to the structural properties of each form, thereby demonstrating the relative advantages and disadvantages of different realization methods presented in this paper.

We derive marginal conditions of optimality (i.e., Euler equations) for a general class of Dynamic Discrete Choice (DDC) structural models. These conditions can be used to estimate structural parameters in these models without having to solve for approximate value functions. This result extends to discrete choice models the GMM-Euler equation approach proposed by Hansen and Singleton (1982) for the estimation of dynamic continuous decision models. We first show that DDC models can be represented as models of continuous choice where the decision variable is a vector of choice probabilities. We then prove that the marginal conditions of optimality and the envelope conditions required to construct Euler equations are also satisfied in DDC models. The GMM estimation of these Euler equations avoids the curse of dimensionality associated to the computation of value functions and the explicit integration over the space of state variables. We present an empirical application and compare estimates using the GMM-Euler equations method with those from maximum likelihood and two-step methods.

This study aims to provide a new method of multiscale directional Lyapunov exponents (MSDLE) calculated based on the state space reconstruction for the nonstationary time series, which can be applied to detect the small target covered by sea clutter.

Reconstructed state space is divided into non-overlapping submatrices whose columns are equal to a predetermined scale. The authors compute eigenvalues and eigenvectors of the covariance matrix of each submatrix and extract the principal components σ_ip and their corresponding eigenvectors. Then, the angles ψ_ip of eigenvectors between two successive submatrices were calculated. The curves of (σ_ip, ψ_ip) reflect the nonlinear dynamics both in kinetic and directional and form a spectrum with multiscale. The fluctuations of (σ_ip, ψ_ip), which are sensitive to the differences of backscatter between sea wave and target, are taken out as the features for the target detection.

The proposed method can reflect the local dynamics of sea clutter and the small target within sea clutter is easily detected. The test on the ice multiparameter imaging X-ban radar data and the comparison to K distribution based method illustrate the effectiveness of the proposed method.

The detection of a small target in sea clutter is a compelling issue, as the conventional statistical models cannot well describe the sea clutter on a larger timescale, and the methods based on statistics usually require the stationary sea clutter. It has been proven that sea clutter is nonlinear, nonstationary or cyclostationary and chaotic. The new method of MSDLE proposed in the paper can effectively and efficiently detect the small target covered by sea clutter, which can be also introduced and applied to military, aerospace and maritime fields.