Find us on Facebook Find us on LinkedIn Follow us on Twitter Subscribe to our YouTube channel Instagram

Business Analytics Seminars

The seminars are on Fridays at 11am in Room 1050, Abercrombie Building (H70), unless otherwise specified.

The seminar organiser is Prof Junbin Gao.

Upcoming Seminars


24th May 2018 - 11:00 am

Venue: Rm 1080 Abercrombie Bldg H70

Speaker: A/Prof Chung-Li Tseng, Operations Management; UNSW Business School

Title: Improving Hand Hygiene Process Compliance through Process Monitoring in Healthcare

We study a compliance problem of healthcare workers (HCWs) in hospitals where hand hygiene compliance rates are generally low. In healthcare, low hand hygiene compliance of HCWs is the leading cause of Hospital-acquired infections, which result in approximately 75,000 patient deaths per year in the US. Using a game-theoretical approach, we model HCWs' reactions to peers' (non-)compliance to determine their equilibrium compliance levels. We integrate the model with a disease-transmission model to determine how the compliance affects disease prevalence in a hospital ward. We establish that a macro-level hand hygiene compliance rate of HCWs can result from a combination of four different types of micro-level non-compliance: free-riding, safe-playing, self-regarding, and opportunistic behaviors.  Finally, we show that the marginal effect of monitoring on reducing disease prevalence depends on clinical factors, HCWs' interpersonal learning, and other integration factors like goal setting. The results demonstrate that the monitoring Intervention may not effectively prevent disease transmission without understanding the micro-level Behaviors of non-compliant HCWs. Our results provide an explanation as to why there is a significant variability in the effectiveness of management intervention as observed in practice.

 


1st Jun 2018 - 11:00 am

Venue: Rm 2290 Abercrombie Bldg H70

Speaker: Prof Manabu Asai, Faculty of Economics; Soka University; Tokyo, Japan

Title: Realized Stochastic Volatility Models with Generalized Asymmetry and Periodic Long Memory

Realized stochastic volatility models of asset returns and realized volatility are considered with a general framework which incorporates higher-order asymmetric function and (possible) periodic long memory. The asymptotic results of a Whittle likelihood estimator are discussed, and Monte Carlo results are presented. An approach to obtain volatility estimates and out-of-sample forecasts are developed. Using high frequency data for three US financial assets, the new model is estimated and evaluated. The forecasting performance of the new model is compared with a realized conditional volatility model.