Speaker: Prof. ZHANG Xinyu, Department of Statistics and Finance, University of Science and Technology of ChinaVenue: Tencent meeting ID: 307-424-966Abstract:Least squares model averaging has been explored in depth from the theoretical perspective and has been used widely in empirical applications. In this talk, I will introduce least squares model averaging and its asymptotic optimality, weight convergence, asymptotic distribution and finite sample property. Also, I will talk about its extension to high-dimensional case, averaging machine learning methods, and non-linear models.

Speakers:GU Shulin, Research Fellow at the Institute of Science and Development, CAS. She also serves as an Advisory Research Professor at the China Institute of Science and Technology Policy (CISTP) at Tsinghua University and Adjunct Professor of the Management School at Zhejiang UniversityILU Xielin, professor of School of Economics and Management, University of Chinese Academy of SciencesWANG Qing, Professor of Marketing and Innovation at Warwick Business School and adjunct Professor at Zhejiang University Management SchoolCHEN Jin, Professor in the School of Economics and Management at Tsinghua University and the Director of the Research Center of Technological Innovation at Tsinghua UniversityJIN Jun, associate professor on innovation management at School of Management, Zhejiang UniversityVenue: Zoom Meeting ID: 996 5200 1716Passcode: 675757Abstract:Chinese scholars recall the working and study experiences with Prof. Freeman, introduce their work on innovation system and development of theories of innovation system in China, and propose their outlooks of research and theories on innovation system in China.See details at:https://mp.weixin.qq.com/s/mNu93drw2JFMu-65tjqRug

Speaker: LIANG Hanying, Professor, School of Mathematical Science, Tongji UniversityVenue: Room 200-9, Run Run Shaw Business Administration Building, Yuquan CampusAbstract:In this talk, we propose estimator of distribution function when the data are right-censored and the censoring indicators are missing at random, and then establish their strong representations and asymptotic normality. Further, based on empirical likelihood method, we define maximum empirical likelihood estimators and smoothed log-likelihood ratios of two-sample quantile difference in the presence and absence of auxiliary information, respectively, and prove their asymptotic distributions. Simulation study and real data analysis are conducted to investigate the finite sample behavior of the proposed methods.

Speaker: Katia Chirkova, researcher, Institute of East Asian Languages, French Academy of Social Sciences, French National Research CenterVenue: Tencent meeting, meeting ID: 836 361 030

Speaker: Rajneesh Narula, John H. Dunning Chair in International Business Henley Business School, Reading UniversityVenue: Zoom（ID：980 5778 6886，Password：466696）Abstract:While COVID-19 has caused significant short-term disruptions in global value chains (GVCs), in the longer run, the pandemic will not be the primary catalyst in GVC evolution. As GVCs recover from the initial shock, managers will make GVC restructuring decisions guided by long-term strategic considerations. We describe barriers that lead firm managers may encounter when rethinking location/control decisions for value chain activities and suggest that, in addition to structural changes, managerial governance adaptations are instrumental in enhancing GVCs’ long-term resilience. Lessons learned from responding to the pandemic can help managers enhance GVC efficiency in the increasingly uncertain global environment.

Speaker: ZHU Fukang, Professor, School of Mathematics, Jilin UniversityVenue: Tencent meeting ID: 788 902 078Abstract:During the last decades, a large variety of models have been proposed for count time series, where the integer-valued autoregressive moving average (ARMA) and integer-valued generalized autoregressive conditional heteroskedasticity (INGARCH) models are the most popular ones. However, while both models lead to an ARMA-like autocorrelation function (ACF), the attainable range of ACF values is much more restricted and negative ACF values are usually not possible. The existing log-linear INGARCH model allows for negative ACF values, but the linear conditional mean and the ARMA-like autocorrelation structure are lost. To resolve this dilemma, a novel family of INGARCH models is proposed, which uses the softplus function as a response function. The softplus function behaves approximately linear, but avoids the drawback of not being differentiable in zero. Stochastic properties of the novel model are derived. The proposed model indeed exhibits an approximately linear structure, which is confirmed by extensive simulations, and which makes its model parameters easier to interpret than those of a log-linear INGARCH model. The asymptotics of the maximum likelihood estimators for the parameters are established, and their finite-sample performance is analyzed via simulations. The usefulness of the proposed model is demonstrated by applying it to three real-data examples.

Speaker: LI Ke, Professor, Harbin Institue of TechnologyVenue: Tencent meeting ID: 873 875 203Abstract:The reliability function, introduced by Shannon in information theory, characterizes the exact exponent under which the error probability of an information processing task approaches zero exponentially. There is little complete result about the reliability function in quantum information. In this talk, I will introduce part of the history and background of this problem, and then I will report our quite recent results on the reliability functions of two quantum information tasks: 1) quantum privacy amplification, and 2) quantum information decoupling. The results are given in terms of a type of information divergence---the Sandwiched Rényi Information Divergence.

Speaker: HAN Pengfei, Assistant Professor, Guanghua School of Management, Peking UniversityTecent Meeting ID: 802 240 481Abstract:We develop measures for technology decoupling and dependence between the U.S. and China based on combined patent data. The first two decades of the century witnessed a steady increase in technology integration (or less decoupling), but China’s dependence on the U.S. increased (decreased) during the first (second) decade. Decoupling in a technology field predicts China’s growing dependence on U.S. technology, which, in turn, predicts less decoupling further down the road. Decoupling is associated with more patent outputs in China, but lower firm productivity and valuation. China’s innovation-oriented industrial policies trade off the inherent conflict between indigenous innovation and firm competitiveness.