題目：Maximum-Likelihood-Like Estimators for the Gamma Distribution
主講人：Zhisheng Ye 博士（National University of Singapore）
時間：2015年6月23日上午10:00
地點：主樓418
主講人簡介：
Dr. Ye Zhisheng is an Assistant Professor in Industrial & Systems Engineering, National University of Singapore, Singapore. He received the B.Eco. in Economics, July 2008, Tsinghua University, Beijing; B.Eng. in Material Science & Engineering, July 2008, Tsinghua University, Beijing; Ph.D. Industrial & Systems Engineering, May 2012, National University of Singapore. His research interests are mainly in reliability modeling, statistical reliability data analysis, etc. The following five papers are his selected publications recently.
1. Ye, Z.S.; Xie, M.; Tang, L.C. and Shen, Y. (2012) Degradation-based burn-in planning under competing risks, Techno metrics, 54 (2), 159-168.
2. Ye, Z.S.; Hong, Y. and Xie, Y (2013) How do heterogeneities in operating environments affect field failure predictions and test planning? The Annals of Applied Statistics, 7(4), 2249-2271.
3. Ye, Z.S. and Ng, K.H.T. (2014) On analysis of incomplete field failure data, The Annals of Applied Statistics, 8(3), 1713-1727.
4. Ye, Z.S.; Xie, M.; Tang, L.C. and Chen, N., (2014) Semi parametric estimation of Gamma processes for deteriorating products, Techno metrics, 56(4), 504-513.
5. Ye, Z.S. and Chen, N. (2014) The inverse Gaussian process as a degradation model, Techno metrics, 56(3), 302-311.
內容簡介：
It is well-known that the maximum likelihood estimators (MLEs) of parameters in a Gamma distribution do not have closed forms. This poses difficulties in some applications such as real-time signal processing using low-grade processors. In this work, we propose two new estimating equations for the gamma distribution. Simple closed-form estimators can be obtained based on these two equations. Performance of the new estimators are shown to be very close to MLEs. Due to the closed form, bias-correction is possible, which significantly improves the small sample performance.
（主辦：管理科學與物流系）
