Bayesian inference for stable Lévy–driven stochastic differential equations with high‐frequency data

In this paper, we consider parametric Bayesian inference for stochastic differential equations driven by a pure‐jump stable Lévy process, which is observed at high frequency. In most cases of practical interest, the likelihood function is not available; hence, we use a quasi‐likelihood and place an associated prior on the unknown parameters. It is shown under regularity conditions that there is a Bernstein–von Mises theorem associated to the posterior. We then develop a Markov chain Monte Carlo algorithm for Bayesian inference, and assisted with theoretical results, we show how to scale Metropolis–Hastings proposals when the frequency of the data grows, in order to prevent the acceptance ratio from going to zero in the large data limit. Our algorithm is presented on numerical examples that help verify our theoretical findings.     

The covariance of the backward and forward recurrence times in a renewal process: the stationary case and asymptotics for the ordinary case

In a Poisson process, it is well-known that the forward and backward recurrence times at a given time point t are independent random variables. In a renewal process, although the joint distribution of these quantities is known (asymptotically), it seems that very few results regarding their covariance function exist. In the present paper, we study this covariance and, in particular, we state both necessary and sufficient conditions for it to be positive, zero or negative in terms of reliability classifications and the coefficient of variation of the underlying inter-renewal and the associated equilibrium distribution. Our results apply either for an ordinary renewal process in the steady state or for a stationary process.     

A Bitcoin-inspired infinite-server model with a random fluid limit

The synchronization process inherent to the Bitcoin network gives rise to an infinite-server model with the unusual feature that customers interact. Among the closed-form characteristics that we derive for this model is the busy period distribution which, counterintuitively, does not depend on the arrival rate. We explain this by exploiting the equivalence between two specific service disciplines, which is also used to derive the model’s stationary distribution. Next to these closed-form results, the second major contribution concerns an asymptotic result: a fluid limit in the presence of service delays. Since fluid limits arise under scalings of the law-of-large-numbers type, they are usually deterministic, but in the setting of the model discussed in this paper the fluid limit is random (more specifically, of growth-collapse type).     

Bayesian nonparametric modelling of the link function in the single-index model using a Bernstein–Dirichlet process prior

This paper proposes the use of the Bernstein–Dirichlet process prior for a new nonparametric approach to estimating the link function in the single-index model (SIM). The Bernstein–Dirichlet process prior has so far mainly been used for nonparametric density estimation. Here we modify this approach to allow for an approximation of the unknown link function. Instead of the usual Gaussian distribution, the error term is assumed to be asymmetric Laplace distributed which increases the flexibility and robustness of the SIM. To automatically identify truly active predictors, spike-and-slab priors are used for Bayesian variable selection. Posterior computations are performed via a Metropolis-Hastings-within-Gibbs sampler using a truncation-based algorithm for stick-breaking priors. We compare the efficiency of the proposed approach with well-established techniques in an extensive simulation study and illustrate its practical performance by an application to nonparametric modelling of the power consumption in a sewage treatment plant.     

Bayesian density regression for discrete outcomes

We develop Bayesian models for density regression with emphasis on discrete outcomes. The problem of density regression is approached by considering methods for multivariate density estimation of mixed scale variables, and obtaining conditional densities from the multivariate ones. The approach to multivariate mixed scale outcome density estimation that we describe represents discrete variables, either responses or covariates, as discretised versions of continuous latent variables. We present and compare several models for obtaining these thresholds in the challenging context of count data analysis where the response may be over‐ and/or under‐dispersed in some of the regions of the covariate space. We utilise a nonparametric mixture of multivariate Gaussians to model the directly observed and the latent continuous variables. The paper presents a Markov chain Monte Carlo algorithm for posterior sampling, sufficient conditions for weak consistency, and illustrations on density, mean and quantile regression utilising simulated and real datasets.     

Every Minute Counts: Using Process Improvement and Performance Feedback to Improve Patient Flow in an Emergency Department

Prolonged emergency department (ED) visits are associated with a number of adverse outcomes for patients as well as lower patient satisfaction scores and increased costs. Several factors that influence the length of ED visits are out of the control of hospital employees, but some opportunities exist to improve performance. For this study, the ED department of a 150-bed hospital in the southeastern United States wanted to improve door-to-discharge time. To do so, a subprocess of door-to-discharge time was targeted, door-to-order. After analyzing the process, the team created standard orders for the 10 most common presenting conditions in the ER with preapproval, allowing nurses to submit the orders without the provider first visiting the patient. Following the process change, daily feedback was added to increase utilization of the preapproved orders. Reductions in door-to-order times and door-to-discharge were observed and patient satisfaction remained stable. Implications for future research in this area are discussed.     

Using Lean Six Sigma to improve mobile order fulfilment process in a telecom service sector

This study proposes and follows a specific and systematic framework for implementing Lean Six Sigma (LSS) methodology in a telecom company in order to improve customer satisfaction by minimizing the company’s response time to customer requirements. The goal of this study was achieved by utilizing several LSS tools under five phases of the DMAIC methodology. Unlike previous studies in the telecom sector that used only qualitative method, in this study, both qualitative and quantitative methods were utilized to draw meaningful conclusions. As a result of the implementation of the LSS methodology, the average order fulfilment lead time for sales orders (SO) and value-added service (VAS) orders was reduced from 10.3 to 5.9 days and from 1.5 to 0.5 days, respectively. The reduction in lead time resulted in an increase in the sigma level for SO and VAS orders from 0.44 to 1.26 and from 0.73 to 2.66, respectively. These improvements were expected to lead to a financial benefit in savings of over $600,000 per year in operational costs, enhancements to customer experience and an increase in revenue generating opportunities. Moreover, this article enriches the existing literature on the application of LSS concept in the service industry, and helps the company to speed up the response to customer requirements.     

目标导向层次分析方法

对含有抽象属性的多属性层次结构而言,层次分析法即AHP（包括DIS-AHP、ABS-AHP、IDE-AHP和SUP-AHP四种具体方法）会因比率比较基准缺失、权重内涵模糊不清或方案评价不保序而缺乏科学理性。为发展AHP,基于摆幅置权（SW）判断模式和多属性决策属性价值公度方法,首先给出了能为层次结构抽象属性上的SW判断提供支持的规约性多属性决策属性价值公度方法,然后由此并结合多属性价值理论给出了能够克服现有层次分析法内在缺陷的目标导向层次分析方法即ToAHP。相对于AHP,ToAHP在判断模式与权重内涵、方法建构的理论基础和相关假设检验、方案评价保序与其内在数理依据上具有明显的相对科学合理性。应用分析表明：在输入信息可比的条件下,ToAHP明显优于AHP的四种分析方法之中最具可信性的SUP-AHP方法。     

基于Hawkes因子模型的股价共同跳跃研究

本文首先比较了三种目前主流的共跳检验方法：基于LM检验的共跳检验、BLT共跳检验和FHLL共跳检验,结果表明,三种方法在识别共跳数量上差距明显,但三者结果的重合部分基本属于市场暴涨暴跌行情,说明共跳识别对市场剧烈波动的聚集性较为敏感。基于跳跃、共跳存在的聚集性问题,本文将Hawkes过程引入跳跃和共跳的研究,构建了基于Hawkes过程的因子模型,结果显示,基于Hawkes因子模型的MJ统计量、CJ统计量和实证数据的拟合程度较好,表明因子模型能够更好地描述跳跃和共跳的聚集性。     

基于Hawkes过程中美股市大幅波动互激效应的研究

近年来,由于中美经济联系日趋紧密,中美股票市场大幅波动的互激效应明显增强。本文考虑中美股市时差和法定节假日差异等因素,运用标值Hawkes过程对2006-2017年CSI300和S&P500大幅波动收益率数据进行建模,结果表明：（1）中美股市大幅波动互激效应存在不对称性,美股市场大幅波动对中国股市的互激效应更强;（2）中美股市大幅波动的幅度对互激效应不存在显著影响;（3）中美股票市场对于大幅波动互激效应的消化速度存在差异,中国股票市场消化美股大幅波动互激效应的速度较快。本研究对金融市场监管者和投资者均有一定意义。