3917 1624
KK 815
Haipeng Shen joined HKU in September 2015 as a Professor of Innovation and Information Management at the Faculty of Business and Economics (now HKU Business School). Before joining HKU, he was a Professor of Statistics and Operations Research, University of North Carolina at Chapel Hill, USA.
Business Analytics, Business Data Analysis, Decision and Risk Analysis, Statistics
Data-driven decision making in the face of uncertainty: big data, business analytics, healthcare analytics, service engineering.
以时间为单位量度客户的耐性并计算危机率，已成为运营管理中重要的一环。这项技术能帮助企业优化客户服务中心的运作，如人手分配及编更事宜等等。当客户申请服务时，他们的耐性将会随时间推移产生改变。现行的数据收集系统有时候无法观察到客户服务中心提供服务的确切时间，故此我们开发了一个TunT（Transform-unTransform）的估计程式，把这个复杂的难题简化为回归分析的问题。我们为客户服务中心不同的服务时间进行分类，并使用均值匹配转换，对相关的数据进行转换，从而使我们能简单地表现出异方差－变异数函数。然后我们把非参数回归分析的技术应用在转换后的数据上。最后，我们会对估计出来的回归函数进行转换，计算出原始危机率。在我们的模拟计算中，我们利用客户服务中心的数据（实验组）与医疗保险计划中的数据（控制组）进行对比。研究证明，与现行方法相比，我们的模型能得出更准确的危机估计值，帮助企业优化人手安排。
Innovation in healthcare is forever changing how we see and experience the medical industry. The environment is offering HKU’s Faculty of Business and Economics (the Faculty) a unique opportunity to be at the forefront of utilising rich data, creating better health outcomes for everyone.
Patients in China suffering from acute ischemic stroke, when arteries leading to the brain are blocked, have traditionally not experienced excellent clinical outcomes. Battling this disease has been a long-term battle for physicians working in the country’s overcrowded under-resourced public hospitals. Professor Haipeng Shen, Patrick S C Poon Professor in Analytics and Innovation at HKU Business School, has been working to change this situation by collaborating with top physicians and embracing the power of big data.
We propose a functional censored quantile regression model to describe the time-varying relationship between time-to-event outcomes and corresponding functional covariates. The time-varying effect is modeled as an unspecified function that is approximated via B-splines. A generalized approximate cross-validation method is developed to select the number of knots by minimizing the expected loss. We establish asymptotic properties of the method and the knot selection procedure. Furthermore, we conduct extensive simulation studies to evaluate the finite sample performance of our method. Finally, we analyze the functional relationship between ambulatory blood pressure trajectories and clinical outcome in stroke patients. The results reinforce the importance of the morning blood pressure surge phenomenon, whose effect has caught attention but remains controversial in the medical literature. Supplementary materials for this article are available online.
Professor Haipeng SHEN, Patrick S C Poon Professor in Analytics and Innovation, was interviewed by four media outlets on his recent research in data-driven decision making in healthcare management.
