May 11, 2015
Sponsored by Center for Statistical Science and
Guanghua School of Management
Venue: Room K02, GSM #2
9:00-9:10 Opening Speech
Li Jin, Guanghua School of Management, Peking University
Kamhom Kan, Institute of Economics, Academia Sinica
George Tiao, Booth School of Business, University of Chicago
Chair: Songxi Chen (Peking University)
9:10-9:40 The Impact of Education on Health
Kamhon Kan, Institute of Economics, Academia Sinica
9:40-10:20 Macroeconomic Forecasting using Approximate Factor Models with Outliers
Ray Y. Chou, Institute of Economics, Academia Sinica and National Chiao Tung University
10:20-10:40 Tea/Coffee Break
Chair: Yundong Tu (Peking University)
10:40-11:20 Association of Cardiovascular Responses with Fine Particle Air Pollutions in Beijing
Jing-Shiang Hwang, Institute of Statistical Science, Academia Sinica
11:20-12:00 Measuring Beijing's PM2.5 Pollution: Severity, Impacts of Weather, APEC and Winter Heating
Songxi Chen, Guanghua School of Management and Center for Statistical Science, Peking University
12:00-14:00 Lunch at Guanghua Dining Hall, B1 Floor
Chair: Xiaojun Song (Peking University)
14:00-14:40 A Unified Approach to Estimating and Testing Income Distributions with Grouped Data
Ying-Ting Chen, Institute of Economics, Academia Sinica
14:40-15:20 Adaptive Quantity Theory of Money
Yundong Tu, Guanghua School of Management and Center for Statistical Science, Peking University
15:20-16:00 Linearity in Mixed-Frequency Regressions
Speaker: Wen-Jen Tsay, Institute of Economics, Academia Sinica
16:00 Closing Remarks by Song Xi Chen
The Impact of Education on Health
Abstract:We investigate the impact of education on health. We use a compulsory education law change in Taiwan to identify the causal effect under the regression discontinuity framework. What is unique about our study is our use of records of medical services utilization to measure an individual's status. Our data are extracted from the 2001--2010 National Health Insurance Research Database and the 2000 Population Census. Our results suggest that education lowers the probability for an individual to have certain illnesses.
Short Bio:Kamhon Kan is current Director of the Institute of Economics, Academia Sinica. After obtaining his Ph.D degree from Virginia Polytechnic Institute and State University in 1993, he spent a year at the Center for Operations Research and Econometrics. He joined Academia Sinica in 1994 and spent a year of sabbatical at the Bank of England in 1996-1997. He research concerns economic analysis of health related issues using microdata. His current interests is the evaluation of education’s causal impact on health outcomes.
Macroeconomic Forecasting using Approximate Factor Models with Outliers
Ray Y. Chou
Abstract:Approximate factor models and their extensions are widely used in forecasting and economic analysis due to their ability in extracting useful information from a large number of relevant variables. In these models, candidate predictors are typically subject to some common components. In this paper, we consider to efficiently estimate an approximate factor model in which the candidate predictors are additionally subject to idiosyncratic large uncommon components such as jumps or outliers. By assuming that occurrences of the uncommon components are rare, we propose an estimation procedure to simultaneously disentangle and estimate the common and uncommon components. We formulate the estimation problem as a penalized least squares problem in which a norm penalty function is imposed on the uncommon components. To solve the estimation problem, we propose an algorithm, which iteratively solves a principal component analysis (PCA) problem and a one dimensional shrinkage estimation problem. The algorithm is flexible in incorporating methods for selecting the number of common components. We then compare finite-sample efficiency of the proposed method and traditional PCA method with simulations. We also demonstrate performances of the proposed method with empirical applications on predicting yearly growths of important macroeconomic indicators.
Short Bio:Ray Y. Chou is a research fellow at the Institute of Economics Academia Sinica and adjunct professor of finance at the National Chiao Tung University. His area of professional expertise is in Financial Econometrics, Asset Pricing and Time Series Forecasting. He has contributed to the use of range-based volatility models in applications to asset allocation and risk management. He holds a BA Degree from National Taiwan University in Economics, a MA in Economics from University of Kentucky and a Ph.D. in Economics from University of California at San Diego (1988). Before his current position, he taught at the Georgia Institute of Technology and was a visiting scholar to the University of Chicago Booth School of Business during 2000-2001. He was also a visiting professor at NYU Stern School and Fordham University during 2013.
Association of cardiovascular responses with fine particle air pollutions in Beijing
Abstract:For concerns about the health of athletes and international visitors to 2008 Olympic Games in Beijing, the government mitigated the ambient air pollution by relocating, limiting or temporarily closing highly polluting, energy-intensive facilities in and around the city, and reducing vehicle usage by elaborate traffic regulations. These air quality interventions, albeit temporary, encouraged numerous investigations on air pollution and its biological effects before, during and after the Games, and provided us a unique opportunity to assess the effect of reduction in fine particles on cardiovascular responses. In this study, Bayesian approaches were used to identify fine particulate matter (PM2.5) sources and estimate their contributions to the ambient air pollution in Beijing. The estimated contributions were brought into mixed-effects models as exposures for examining the association of cardiovascular responses of the exposed mice in a sub-chronic experiment. We will show how the alterations in cardiac parameters were closely related to changes in Beijing ambient PM2.5 concentration and various pollution source concentrations.
Short Bio:Dr. Jing-Shiang Hwang is a Distinguished Research Fellow of the Institute of Statistical Science, Academia Sinica in Taiwan. He is also a faculty member of the Institute of Public Health at National Yang-Ming University and Institute of Public Health at National Cheng Kung University. He received his PhD in Statistics from Harvard University in 1992. He has been working on applied statistics at Academia Sinica since then. His current research interests include developing methods for environmental health research, cost-effectiveness analysis and social network analysis.
Measuring Beijing's PM2.5 Pollution: Severity, Impacts of Weather, APEC and Winter Heating
Abstract:By learning from five years' PM2.5 and meteorological data, the severity of PM2.5 pollution in Beijing is quantified with a set of statistical measures. As PM2.5 concentration is highly influenced by meteorological conditions, we propose a statistical approach to adjust PM2.5 concentration with respect to meteorological conditions, which can be used to monitor PM2.5 pollution in a location. The adjusted monthly averages and percentiles are employed to test if the PM2.5 levels in Beijing have been lowered since the China's State Council setting up pollution reduction target. The results of the testing reveals significant increases, rather than decreases, in the PM2.5 concentrations in Years 2013 and 2014 as compared to those in Year 2012, respectively. We conduct analyses on two quasi-experiments: the Asia-Pacific Economic Cooperation (APEC) meeting in November 2014 and the annual winter heating, to gain insight on the impacts of emissions on PM2.5. The analyses lead to a common conclusion that a fundamental shift from the mainly coal-based energy consumption to more greener alternatives in Beijing and the surrounding North China Plain is the key to solve the PM2.5 problem in Beijing.
Co-Authors: Xuan Liang, Tao Zou, Bin Guo, Shuo Li, Haozhe Zhang, Shuyi Zhang, Hui Huang
A Unified Approach to Estimating and Testing Income Distributions with Grouped Data
Abstract:In studying income distributions, it is common that researchers can only access certain types of grouped data. In this paper, we propose a unified approach which is flexibly applicable to estimating and testing parametric income distributions in a general context of grouped data. This context includes, but is not limited to, the relative frequency table and the income share table. In addition, we also provide a parametric bootstrap method and show its validity for statistical inference. We also compare our methods with existing methods for grouped income data, and evaluate the finite-sample performance of the proposed methods by a Monte Carlo simulation.
In the empirical part, we apply our methods to exploring the time evolution of China's income/consumption distribution from a collection of income share tables.
JEL classification: C22, C52, D31.
Keywords:Bootstrap, GMM, Grouped data, Income distribution, Over-identifying restriction test.
Short Bio:Professor Chen received his Ph.D. in Economics from National Taiwan University in 1998. He is currently a Research Fellow at the Institute of Economics, Academia Sinica. His primary research fields are Econometrics and Empirical Finance. He has publications in several journals on Econometrics. His current research interests include topics on income distribution, nowcasting and forecasting business conditions, model specification test, model uncertainty and program evaluation.
Adaptive Quantity Theory of Money
Abstract:In this article we investigate the link between price level (P), money supply (M) and real output (Y) in the U.S. since 1959. We find that price (log(P)) is no longer in proportion with excess money supply, as measured by log(M/Y ), after 2000, in contrast to what is postulated by the Quantity Theory of Money (QTM). This relationship is found unstable and depends closely on macroeconomic state variables such as the rate of unemployment. As a remedy, we generalize the notion of cointegration and allow the cointegration vector to adapt with unemployment rate, thus obtain an Adaptive Quantity Theory of Money (AQTM). AQTM is shown to more precisely describe the data and it is robust to model specification tests. Formulated in the form of Engle and Granger's (1987) error correction model (ECM), we found that AQTM outperforms the traditional QTM and other models used frequently in forecasting U.S. inflation. This is a joint work with Song Xi Chen and Ying Wang.
Linearity in Mixed-Frequency Regressions
Short Bio:Wen-Jen Tsay is a Research Fellow of Acadmia Sinica. He received Ph.D. in Economics from Michigan State University in 1995. Her research interests are in International Finance, Macroeconometrics, Stochastic Frontier Analysis, and Electoral Studies. He served as referees for over 30 international journals. He published several papers in Journal of Econometrics, Econometric Theory, Journal of Time Series Analysis, Electoral Studies, Journal of Population Economics, and Journal of Productivity Analysis, etc.