Prediction of hotel bankruptcy using support vector machine,artificial neural network,logistic regression,and multivariate discriminant analysis

The objectives of this paper are firstly, to provide an optimal hotel bankruptcy prediction approach to minimize the empirical risk of misclassification and secondly, to investigate the functional characteristics of multivariate discriminant analysis, logistic, artificial neural networks (ANNs), and support vector machine (SVM) models in hotel bankruptcy prediction. The performances were evaluated not only in terms of overall classification and prediction accuracy but also in terms of relative error cost ratios. The results showed that ANN and SVM were very applicable models in bankruptcy prediction with data from Korean hotels. When jointly measuring both type I and type II errors, especially allowing for the greater costs associated with type I errors, however, ANN was more accurate with smaller estimated relative error costs than SVM. Thus, if the objective is to find the best early warning technique that performs accurately with small relative error costs, then, it will be worth considering ANN method for hotel bankruptcy prediction.     

Can Industry Consolidation Lead to Greater Efficiencies? Evidence from the U.S. Defense Industry

The question of whether mergers in various industries lead to greater market power or improved efficiencies has been the subject of numerous public policy debates. This analysis focuses on the impact of consolidation in the U.S. defense industry over the past 20 years and examines the reasons behind the wave of defense consolidation, the results in terms of the reduction in contractors, the antitrust response to mergers, and evidence on the impact of the mergers on weapons systems’ total and per-unit costs. The analysis finds that merger activity was driven less by declines in spending following the Cold War than by a stronger economy and a vibrant financial market. The cost data show that 39 to 44 percent of systems experienced statistically significant change in either total costs or per-unit costs following a merger. Somewhat more systems were likely to exhibit lower postmerger per-unit costs than higher per-unit costs, suggesting improved efficiency. The analysis also examines the impact on weapon systems cost by type of weapons system, manufacturer, and service (Army, Navy, Air Force). The evidence suggested greater efficiencies following consolidation for many sectors. Army and Navy systems overall showed lower per-unit costs, but the Air Force weapons systems showed mixed results.     

Medicaid, State Finances, and the Bottom Line for Businesses

Current Medicaid expenditures account for about nine percent of the federal budget and almost a quarter of state budgets and are growing rapidly. State Medicaid budgets are especially vulnerable to recession since states cannot incur large and sustained fiscal deficits. Without change, Medicaid burdens will cause state finances to be diverted from infrastructure and education, with negative effects on the costs and productivity of business. Also affecting business are the state governments experimenting with policies that shift Medicaid burdens to private employers. Simultaneously, the states are initiating efforts to ease Medicaid’s relentless cost increases and address its long-run problems. JEL Classification I180,H720     

Artificial Neural Systems as a Research Paradigm for the Study of Group Decision Support Systems

Artificial neural systems (ANSs) have received interest recently because of their ability to accurately forecast and classify data. Group decision support systems (GDSSs) also have received much interest for their support of group communication and decision making. This paper explores the potential of the ANS as a methodology for modeling the many complex, interrelated research variables involved in the field of GDSS. As an illustration, multilinear regression, an ANS with backpropagation, and an ANS with a genetic algorithm were developed to classify 133 subjects into verbal or GDSS groups based on their responses to a questionnaire. The ANSs with backpropagation and the genetic algorithm achieved higher classification accuracies (81.8 and 90.9%, respectively) than was achieved with multilinear regression (75.8%). Therefore, an ANS may more accurately model the many interrelationships occurring with GDSS group behavior.     

Strategic investment analysis for migration beyond 3G wireless networks

This paper analyzes investment alternatives for cellular operators who deploy either universal mobile telecommunications systems (UMTS) with high speed data packet access (HSDPA) networks or UMTS with Wi-Fi networks, for moving beyond 3G-networks. The evaluated business scenarios focus on two different deployment alternatives: UMTS with HSDPA and UMTS integrated with Wi-Fi. HSDPA deployment targets the overlay on the existing UMTS cells, whereas Wi-Fi covers only indoor hot spot areas where data demand is high. Capital investments and operational costs as well as revenue streams have been analyzed for the two types of network architectures. Techno-economic methodologies have been applied for this study and results are derived that are presented and discussed for two types of wireless networks.     

Development of crash frequency models for safety promotion of urban collector streets

The merits for development and application of crash frequency prediction models for safety promotion on any road type, with a focus on urban collector streets, are presented in this article. The city of Yazd, a medium-sized city in the middle of Iran, was selected as a case study and the data required for modelling crash frequencies along five collector streets comprising 31 street sections were collected. Six models including Poisson and negative binomial models and their deviations along with a hybrid artificial neural networks (ANN) model were developed to predict crash frequency along each street section. The overfitting problem was addressed using appropriate sensitivity analysis methods which were also used to identify the input variables with significant impact on the model performance. The results indicated that the developed hybrid ANN model provided the best performance in terms of accuracy and the number of input variables. The application of hybrid ANN model to evaluate the safety impacts of four different strategies, each resembled by one of the input variables of this model, indicated that these models can successfully be used for this purpose.     

ASSESSING THE VIABILITY OF REUSABLE PACKAGING: A RELATIVE COST APPROACH

While environmentally friendly operations can provide a positive public image, firms wishing to reduce their environmental impact must initiate activities that reduce cost as well. Reusable container systems are one means of reducing environmental waste, but the economics of such systems has been somewhat uncertain to date. This research proposes a cost model to evaluate the combined impact of logistics and packaging costs on the container decision. Additionally, a simulation/regression analysis is used to analyze model results to determine the relative importance and interactions between container decision factors.     

Bottom-up framework for cost allocation to services in telecommunication networks

Communications service providers (CSP) supply information, transport and passive infrastructure services that rely on the deployment of multiservice networks. To attain service cost reductions, the planning and design of those networks is essential, so that the total cost of the network ownership (TCO) is minimized. In this paper, we propose a bottom-up calculation framework for the allocation of the TCO cost components (i.e. capital and operational expenditures) to the services supplied by a CSP. For this, the framework is formulated within the context of the network life-cycle. Our method provides the level of abstraction required for the calculation of the three basic service cost contributors, namely, direct, shared and common costs. Moreover, it outlines the role of network planning in the design of an efficient network infrastucture, and in the calculation of the TCO. Based on this, we show how the stand alone, fully allocated and long-run incremental costs of a service can be calculated. Besides the determination of service cost drivers, the framework is useful to evaluate the cost-efficiency of competing technologies. To illustrate this, we study the allocation of the infrastructure costs for a CSP offering transport services with an optical network. By designing the infrastructure with fixed-grid and flex-grid technologies, it is shown that in both cases the main service cost driver is determined by the transponders installed. The results show that fixed-grid fits well for the provisioning of high speed constant bit-rate transport services, whereas flex-grid is a good candidate for transport services with bursty traffic patterns.     

Costs and Benefits of Recycling Liquid Board Containers

Recycling of packaging material has become more or less mandatory in many European countries, including Norway. Through so-called voluntary agreements quantitative targets are set for the proportion of total waste to be recycled. At the same time the strategic objective for Norwegian waste policy is that there should be a socio-economic balance between different waste treatment options. On the basis of a cost-benefit analysis it is questioned whether the Norwegian recycling policy for liquid board containers really is cost-effective. The calculations show that the net social costs of the recycling system in 1999 amounted to EUR3.5 million. The high cost is due to the fact that these containers constitute a small fraction of total waste from the household and that it is costly to separate it from other waste. The environmental costs from landfilling or incineration are small compared to the costs of recycling. The best alternative, according to our analysis, is to incinerate the containers with energy recovery.     

微蜂窝中电波传播预测模型的改进

在传统的电波传播预测模型的基础上，提出了一种基于射线跟踪法的改进模型。该模型可真正用于除圆形建筑物以外的多棱角建筑物组成复杂微蜂窝区域中场强的准确预测，具有高自适应性、预测精度可有需求控制和高效率等优点。对该模型所得预测结果同文献进行了比较，结果表明该模型具有较高效率和精确度。     

Environmental reporting: Toward enhanced information quality

Managers and stakeholders are increasingly aware of the importance of the environmental impact of a firm's operations when assessing risk and attempting to determine future profitability. Unfortunately, financial accounting systems often fail to fully disclose these environmentally-related costs. The reasons underlying this incomplete disclosure are myriad, ranging from measurement issues to the structure of the firm's chart of accounts. In many ways, the issues facing managers and stakeholders who are attempting to assess environmental costs arising from business operations resemble the issues faced when attempting to determine the costs of producing poor quality products. The negative impact on the environment from business operations can be viewed as a failure in the same way that the negative impact of producing a defective product can be seen as a production control failure. Similarly, costs are incurred to prevent and detect environmental failures, and the cost of failure—particularly if not addressed within the firm—can be huge and unknowable. Drawing on the experiences of firms employing quality measures and reporting, this article presents an environmental cost reporting model to provide greater transparency on environmental impact of business operations to managers and firm stakeholders.     

Estimating the costs of non-fatal consumer product injuries in the United States

This paper describes a data-driven injury cost model (ICM) developed to estimate the costs associated with non-fatal consumer product injuries. The modeling effort combines information by diagnosis from the US Consumer Product Safety Commission’s National Electronic Injury Surveillance System (NEISS) and 17 other large data sets. The ICM contains four aggregated cost components: (1) medical costs, (2) work losses, (3) quality of life and pain and suffering costs, and (4) product liability insurance administration and litigation costs. The ICM estimates societal costs, which are broader than costs to any individual group, such as victims, insurers, or product manufacturers. Costs associated with consumer product injuries are estimated to be approximately $500 billion in 1996, accounting for nearly one-third of the total annual injury costs. We examine injury costs in several ways, including by major product category, by sex and age of victims, by body part injured, by injury diagnosis, and by highest level of medical treatment received. We also rank the 10 leading consumer products that account for injury costs overall and within different age groups. Products such as stairs and floors are among the top 10 for all age groups. Other products, however, are more closely tied to injuries at particular stages of life (e.g., infant/toddler, child, young adult, elderly). These cost estimates are useful in assessing which products and types of injuries impose the greatest costs on society and for identifying areas for focused injury prevention efforts.     

Clustering-based classification of road traffic accidents using hierarchical clustering and artificial neural networks

Artificial neural networks (ANNs) have been widely used in predicting the severity of road traffic crashes. All available information about previously occurred accidents is typically used for building a single prediction model (i.e., classifier). Too little attention has been paid to the differences between these accidents, leading, in most cases, to build less accurate predictors. Hierarchical clustering is a well-known clustering method that seeks to group data by creating a hierarchy of clusters. Using hierarchical clustering and ANNs, a clustering-based classification approach for predicting the injury severity of road traffic accidents was proposed. About 6000 road accidents occurred over a six-year period from 2008 to 2013 in Abu Dhabi were used throughout this study. In order to reduce the amount of variation in data, hierarchical clustering was applied on the data set to organize it into six different forms, each with different number of clusters (i.e., clusters from 1 to 6). Two ANN models were subsequently built for each cluster of accidents in each generated form. The first model was built and validated using all accidents (training set), whereas only 66% of the accidents were used to build the second model, and the remaining 34% were used to test it (percentage split). Finally, the weighted average accuracy was computed for each type of models in each from of data. The results show that when testing the models using the training set, clustering prior to classification achieves (11%–16%) more accuracy than without using clustering, while the percentage split achieves (2%–5%) more accuracy. The results also suggest that partitioning the accidents into six clusters achieves the best accuracy if both types of models are taken into account.     

Heuristic algorithms for a second-best congestion pricing problem

Designing a congestion pricing scheme involves a number of complex decisions. Focusing on the quantitative parts of a congestion pricing system with link tolls, the problem involves finding the number of toll links, the link toll locations and their corresponding toll level and schedule. In this paper, we develop and evaluate methods for finding the most efficient design for a congestion pricing scheme in a road network model with elastic demand. The design efficiency is measured by the net social surplus, which is computed as the difference between the social surplus and the collection costs (i.e. setup and operational costs) of the congestion pricing system. The problem of finding such a scheme is stated as a combinatorial bi-level optimization problem. At the upper level, we maximize the net social surplus and at the lower level we solve a user equilibrium problem with elastic demand, given the toll locations and toll levels, to simulate the user response. We modify a known heuristic procedure for finding the optimal locations and toll levels given a fixed number of tolls to locate, to find the optimal number of toll facilities as well. A new heuristic procedure, based on repeated solutions of a continuous approximation of the combinatorial problem is also presented. Numerical results for two small test networks are presented. Both methods perform satisfactorily on the two networks. Comparing the two methods, we find that the continuous approximation procedure is the one which shows the best results.     

降低企业物流成本

目前,我国物流企业费用高、库存成本及运输成本高,物流信息技术落后是不争的事实。其原因是会计核算不健全,物流成本管理缺乏综合控制手段。降低物流企业成本,应科学规划物流节点,提高物流合理化水平,采用供应链方式追求成本效益,加快物流运行速度,提高资金利用效率,不断提高物流服务质量,构建现代化物流网络和信息系统,增强物流企业成本控制能力。     

INTERNATIONAL RESERVE MANAGEMENT: A DRIFT‐SWITCHING REFLECTED JUMP‐DIFFUSION MODEL

We study the cost of shocks, that is, jump risk, with respect to reserve management when the reserve process is formulated as a drift‐switching jump diffusion with a reflecting barrier at 0. Inspired by the Brownian drift switching model, our model results in a more realistic dynamic behavior of international reserves than the buffer stock model. The new model can capture both the jump behavior in reserve dynamics and the leptokurtic feature of the increment distribution which has a higher peak and two asymmetric heavier tails than the normal distribution. Through the selection of an initial distribution that reflects certain steady state behaviors, the reserve process becomes a regenerative process. This selection enables us to derive a closed‐form expression for the total expected discounted cost of managing reserves, thus helping us to numerically find management strategies that minimize costs. The numerical results show that shocks at the reserve level have a significant effect on reserve management strategies and that model misspecification can result in nonnegligible additional costs.     

Sunk Costs and the Entry Decision

This paper considers the role of sunk costs in the decision to enter a market. Its goal is to provide a policy-relevant approach to the question: when are sunk costs so great as to serve as a barrier to entry? To do this, the model presented nests both a model of pure hit-and-run entry and a simple “lottery-ticket” model of long-term entry in which the entrant knows that entry may or may not ultimately prove successful. It illustrates clearly the strategic differences between sunk and nonsunk fixed costs. The paper also considers the incumbent’s problem of choosing between entry deterrence and accommodation. Finally, out of this model comes a measure of the height of the sunk cost barrier to entry that may be useful for competition policy purposes.     

Franchising as a plural system: A risk-based explanation

Empirical studies show that most franchise systems consist of both franchisee-owned and franchisor-owned units. We contribute a new theory that explains why such a mixture exists, using a model that focuses on the franchisor's optimal risk allocation. The costs of risk and controlling franchised units explain the varying fraction of franchisee-owned to total selling units, and the incentive to franchise decreases with an increasing fraction of franchisee-owned to total selling units, as well as with decreasing costs of control. Our explanation for these plural systems is consistent with the ownership redirection hypothesis.     

Laffont and Tirole's Competition in Telecommunications : A View From the US

This article briefly reviews Competition in Telecommunications by Laffont andTirole (2000); it presents the major conclusions of the book and corrects certain errors. In addition, this article treats related topics, some specific to the US telecommunications industry. In US telecommunications access to incumbents facilities are priced based on cost model estimates. Several of the important sources of misunderstandings of telecommunications costs and cost estimates are described including: multiple meanings of 'access,'meanings of cross-subsidy; nature of loop costs; the many dimensions of marginal costs in telecommunications, misuses of long-run cost concepts; and misuses of 'most efficient' provider assumptions. The methods of cost calculations and legal and regulatory requirements in the US are described, as are the implications for incentive regulation, changing technology, efficient pricing, and measures of efficient competitive entry.