Global shipment volumes have been increasing due to changes in the business environment of e-commerce and manufacturing. Consequently, container vessels carry more cargo for international trade, increasing uncertainti...
详细信息
Global shipment volumes have been increasing due to changes in the business environment of e-commerce and manufacturing. Consequently, container vessels carry more cargo for international trade, increasing uncertainties in terminal management. Terminal operators manage terminals by establishing a proactive schedule that responds to disruptions such as vessel delays, and the introduction of buffer time is a representative proactive strategy. In this study, by analyzing historical delay data with machine learning, we propose data-driven buffer times to consider the heterogeneous arrival uncertainty of vessels. Thus, we proactive scheduling with data-driven buffer times according to the desired robustness levels. This is a novel study on berth scheduling that applies data mining approaches to improve operations research techniques. Numerical experiments were conducted on the berth scheduling with time-invariant quay crane assignment using real-life data to validate the effectiveness of the proposed method. These experimental results revealed that applying the data-driven buffer time could effectively reduce the cost incurred at the terminal by balancing baseline and recovery costs. In addition, our proposed methodology ensured the quality of the solution compared with a stochastic method and reduced the computational burden of a stochastic problem by using the data-driven buffer times obtained prior to the solution construction. Therefore, the proposed method can be introduced into terminal operations to overcome the deficiencies of traditional approaches in terms of academic perspective.
The mixture modeling framework is widely used in many applications. In this paper. we propose it component reduction technique, that collapses it Gaussian mixture model into a Gaussian mixture with fewer components. T...
详细信息
The mixture modeling framework is widely used in many applications. In this paper. we propose it component reduction technique, that collapses it Gaussian mixture model into a Gaussian mixture with fewer components. The em (Expectation-Maximization) algorithm is usually used to fit a mixture model to data. Our algorithm is derived by extending mixture model learning using the em-algorithm, In this extension. a difficulty arises from the fact that some crucial quantities cannot be evaluated analytically. We overcome this difficulty by introducing an effective approximation. The effectiveness of our algorithm is demonstrated by applying it to a simple synthetic component reduction task and a phoneme clustering problem.
We introduce a broad and flexible class of multivariate distributions obtained by both scale and shape mixtures of multivariate skew-normal distributions. We present the probabilistic properties of this family of dist...
详细信息
We introduce a broad and flexible class of multivariate distributions obtained by both scale and shape mixtures of multivariate skew-normal distributions. We present the probabilistic properties of this family of distributions in detail and lay down the theoretical foundations for subsequent inference with this model. In particular, we study linear transformations, marginal distributions, selection representations, stochastic representations and hierarchical representations. We also describe an em-typeal gorithm for maximum likelihood estimation of the parameters of the model and demonstrate its implementation on a wind dataset. Our family of multivariate distributions unifies and extends many existing models of the literature that can be seen as submodels of our proposal. (C) 2018 Elsevier Inc. All rights reserved.
Inverse Gaussian distribution has been used widely as a model in analysing lifetime data. In this regard, estimation of parameters of two-parameter (IG2) and three-parameter inverse Gaussian (IG3) distributions based ...
详细信息
Inverse Gaussian distribution has been used widely as a model in analysing lifetime data. In this regard, estimation of parameters of two-parameter (IG2) and three-parameter inverse Gaussian (IG3) distributions based on complete and censored samples has been discussed in the literature. In this paper, we develop estimation methods based on progressively Type-II censored samples from IG3 distribution. In particular, we use the em-algorithm, as well as some other numerical methods for determining the maximum-likelihood estimates (MLEs) of the parameters. The asymptotic variances and covariances of the MLEs from the em-algorithm are derived by using the missing information principle. We also consider some simplified alternative estimators. The inferential methods developed are then illustrated with some numerical examples. We also discuss the interval estimation of the parameters based on the large-sample theory and examine the true coverage probabilities of these confidence intervals in case of small samples by means of Monte Carlo simulations.
Statistical analysis often involves the estimation of a probability density based on a sample of observations. A commonly used nonparametric method for solving this problem is the kernel-based method. The motivation i...
详细信息
Statistical analysis often involves the estimation of a probability density based on a sample of observations. A commonly used nonparametric method for solving this problem is the kernel-based method. The motivation is that any continuous density can be approximated by a mixture of densities with appropriately chosen bandwidths. In many practical applications, we may have specific information about the moments of the density. A nonparametric method using a mixture of known densities is proposed that conserves a given set of moments. A modified expectation-maximisation algorithm for estimating the weights of the mixture density is then developed. The proposed method also obtains an estimate of the number of components in the mixture needed for optimal approximation. The proposed method is compared with two popular density estimation methods using simulated data and it is shown that the proposed estimate outperforms the others. The method is then illustrated by applying it to several real-data examples.
In this paper, we discuss inferential aspects for the Grubbs model when the unknown quantity x (latent response) follows a skew-normal distribution, extending early results given in Arellano-Valle et al. (J Multivar A...
详细信息
In this paper, we discuss inferential aspects for the Grubbs model when the unknown quantity x (latent response) follows a skew-normal distribution, extending early results given in Arellano-Valle et al. (J Multivar Anal 96:265-281, 2005b). Maximum likelihood parameter estimates are computed via the em-algorithm. Wald and likelihood ratio type statistics are used for hypothesis testing and we explain the apparent failure of the Wald statistics in detecting skewness via the profile likelihood function. The results and methods developed in this paper are illustrated with a numerical example.
An iterative procedure for estimating HIV incidence from AIDS incidence data is discussed. The approach. proposed by Becker et al. (1991), makes use of the emalgorithm combined with a smoothing step. In this paper we...
详细信息
An iterative procedure for estimating HIV incidence from AIDS incidence data is discussed. The approach. proposed by Becker et al. (1991), makes use of the emalgorithm combined with a smoothing step. In this paper we suggest a modification of the method, based on an alternative choice for the complete data model used to invoke the emalgorithm. It is found that this modification improves the stability and flexibility of the estimates, although it generally leads to slower convergence.
The evolution of DNA sequences can be described by discrete state continuous time Markov processes on a phylogenetic tree. We consider neighbor-dependent evolutionary models where the instantaneous rate of substitutio...
详细信息
The evolution of DNA sequences can be described by discrete state continuous time Markov processes on a phylogenetic tree. We consider neighbor-dependent evolutionary models where the instantaneous rate of substitution at a site depends on the states of the neighboring sites. Neighbor-dependent substitution models are analytically intractable and must be analyzed using either approximate or simulation-based methods. We describe statistical inference of neighbor-dependent models using a Markov chain Monte Carlo expectation maximization (MCMC-em) algorithm. In the MCMC-emalgorithm, the high-dimensional integrals required in the emalgorithm are estimated using MCMC sampling. The MCMC sampler requires simulation of sample paths from a continuous time Markov process, conditional on the beginning and ending states and the paths of the neighboring sites. An exact path sampling algorithm is developed for this purpose.
This paper develops parametric inference for the parameters of location-scale family of distributions based on a ranked set sample. Likelihood function incorporates within-set ranking errors into the model through a m...
详细信息
This paper develops parametric inference for the parameters of location-scale family of distributions based on a ranked set sample. Likelihood function incorporates within-set ranking errors into the model through a missing data mechanism. The maximum likelihood estimators of the location-scale and missing data model parameters are constructed and an em-algorithm is provided. It is shown that the proposed estimator is robust against imperfect ranking error and provides higher efficiency over its competitors. (C) 2010 Elsevier B.V. All rights reserved.
Envelope method was recently proposed as a method to reduce the dimension of responses in multivariate regressions. However, when there exists missing data, the envelope method using the complete case observations may...
详细信息
Envelope method was recently proposed as a method to reduce the dimension of responses in multivariate regressions. However, when there exists missing data, the envelope method using the complete case observations may lead to biased and inefficient results. In this paper, we generalize the envelope estimation when the predictors and/or the responses are missing at random. Specifically, we incorporate the envelope structure in the expectation-maximization (em) algorithm. As the parameters under the envelope method are not pointwise identifiable, the emalgorithm for the envelope method was not straightforward and requires a special decomposition. Our method is guaranteed to be more efficient, or at least as efficient as, the standard emalgorithm. Moreover, our method has the potential to outperform the full data MLE. We give asymptotic properties of our method under both normal and non-normal cases. The efficiency gain over the standard em is confirmed in simulation studies and in an application to the Chronic Renal Insufficiency Cohort (CRIC) study.
暂无评论