您的位置是:首页 > 新闻公告 >新闻活动

【YEAP】January 14-15th, 2017

2017-01-05


2017 Meeting of Young Econometricians

in Asia-Pacific (YEAP) Region

 

 

January 14-15, 2017 

 

Sponsors:      National Academy of Development and Strategy

Renmin University of China

 

               Institute of Statistics and Big Data,

Renmin University of China

 

               Center of Quantitative Investment, Hanqing Advanced Institute of Economics and Finance, Renmin University of China

 

 

Organizer:  National Academy of Development and Strategy

Renmin University of China

 

Institute of Statistics and Big Data

Renmin University of China

 

 

Scientific Committee

 

Chunrong Ai      Institute of Statistics and Big Data, Renmin University of China

Yong Li          Hanqing Advanced Institute of Economics and Finance,

Renmin University of China

Jun Ma           School of Economics, Renmin University of China

Liping Zhu        Institute of Statistics and Big Data, Renmin University of China

Yonghui Zhang     School of Economics, Renmin University of China

   

 

 

Organizing Committee

 

Yong Li (Chair)         Hanqing Advanced Institute of Economics and Finance

Liping Zhu (Chair)      Institute of Statistics and Big Data

Lijie Zhang (Secretary)   Hanqing Advanced Institute of Economics and Finance 

                      Tel: 18813192540

Liu He (Secretary)       Institute of Statistics and Big Data

                      Tel: 18511808238


 

Conference Program

Location: Room 123, School of Chinese Classics

Renmin University of China

                        中国人民大学国学馆123

 

Saturday, Jan 14, 2017

08:00-08:40  Registration

08:40-09:00  Opening Ceremony

09:00-10:00

Keynote Speech I

Title: A Simple Method of Estimating Nonparametric Conditional Quantile Function

Speaker: Qi Li (Texas A&M University & Capital University of Economics and Business)

Chair: Liping Zhu (Renmin University of China)

10:00-10:30   Photo Taking & Tea/Coffee Break

Section I:   Nonparametric Testing

 

Chair: Yundong Tu (Peking University)

10:30-11:00

Nonparametric Test of Monotonicity of Bidding Strategy in First-price Auctions (Nianqing Liu and Quang Vuong)

Speaker: Nianqing Liu (Shanghai University of Finance and Economics)

11:00-11:30

Nonparametric Measuring and Testing of Heteroskedasticity (Xiaojun Song and Abderrahim Taamouti)

 

Speaker: Xiaojun Song (Peking University)

11:30-12:00

A Simultaneous Test for Model Specification of Nonlinear Time Series Models (Shou Li, Bin Guo and Yundong Tu)

 

Speaker: Shou Li (Tianjin University of Finance and Economics)

12:00-13:30  Lunch

Section II:  Bayesian Econometrics

 

Chair: Xiaojun Song (Peking University)

14:00-14:30

Bayesian Specification Testing for Latent Variable Models (Yong Li, Tao Zeng and Jun Yu)

Speaker: Tao Zeng (Wuhan University)

14:30-15:00

Functional Coefficient Time Series Models with Trending Regressors

Speaker: Tingting Cheng (Nankai University)

15:00-15:30

A New Bayesian Estimator for Life-Cycle Model

Speaker: Xiaobin Liu (Singapore Management University)

15:30-16:00

Tea/Coffee Break

Section III Large Panel Data Models

 

Chair: Xinyu Zhang (Chinese Academy of Sciences)

16:00-16:30

Quantile Factor Models (Liang Chen, Juan J. Dolado, and Jesús Gonzalo)

Speaker: Liang Chen (Shanghai University of Finance and Economics)

16:30-17:00

Specification Tests for Large Dynamic Panel Data Models (Yoonseok Lee and Bin Peng)

Speaker: Bin Peng (Huazhong University of Science and Technology)

17:30-18:00

Dynamic Factor VaR Measurement for Large Portfolios in an Emerging Marker

Emerging Market (Yanyan Guo and Yanyun Zhao)

Speaker: Yanyun Zhao (Zhongnan University of Economics and Law)

18:00-20:00  Reception Dinner (by invitation, Discussion for next YEAP conference )

 

 

20:00-22:00  Discussion Section on the progress and challenges of DSGE models  (by invitation)  

 

Sunday, Jan 15, 2017

09:00-10:00

Keynote Speech II

Title: Does China Grow by Sending More Students to College

Speaker: Shuaizhang Feng (Jinan University)

Chair: Yong Li (Renmin University of China)

Section IV:  Econometric Applications

 

Chair: Yonghui Zhang (Renmin University of China)

10:30-11:00

Forecasting with Social Media Data in the Presence of Heteroscedasticity (Steven Lehrer and Tian Xie)

Speaker: Tian Xie (Xiamen University)

11:00-11:30

Chinese Trade Price and Yuan`s Valuation

Speaker: Yichen Gao (Capital University of Economics and Business)

11:30-12:00

The Impact of Export Expansion on Youths Education: Triple Difference Evidence from Chinas WTO Accession

 

Speaker: Faqin Lin (Central University of Finance and Economics)

12:00-13:30

Lunch

Section V:  Financial Econometrics

 

Chair: Tian Xie (Xiamen University)

14:00-14:30

Adaptive Estimation of Functional-Coefficient Cointegration Models with Nonstationary Volatility

Speaker: Ying Wang (Peking University)

14:30-15:00

Portfolio Analysis of Skewed Data with Risk Measure (Ruili Sun and Tiefeng Ma)

Speaker: Ruili Sun (Southwestern University of Finance and Economics)

15:00-15:30

Conditional Covariance Model

Speaker: Jin Liu (Peking University)

15:30-16:00  Closing Ceremony


Abstract

Section I:   Nonparametric Testing

 

Nonparametric Test of Monotonicity of Bidding Strategy in First-Price Auction. (Nianqing Liu and Quang Vuong)

Speaker: Nianqing Liu (Shanghai University of Finance and Economics)

 

This paper develops a nonparametric test of monotonicity of bidding strategy in first price auctions. As shown by Guerre, Perrigne, and Vuong (2000), monotonicity of bidding strategy is the essential restriction imposed by the (theoretical) symmetric first price auctions on distribution of bids. Based on the equivalence between monotonicity of bidding strategy and convexity of integrated value quantile of a bidder’s strongest competitor, we propose a test statistic measuring a distance of the integrated value quantile of a bidder’s strongest competitor from convexity. It only involves in estimation of bid quantile function, and hence avoids smoothing estimation of bid density. Our test with bootstrap critical values is shown to have the correct size asymptotically, to be consistent against all fixed alternatives and to have non-trivial power against root-n local alternatives. Monte Carlo experiments show that our testing procedure works well in finite samples.

 

 

Nonparametric Measuring and Testing of Heteroskedasticity (Xiaojun Song and Abderrahim Taamouti)

Speaker: Xiaojun Song (Peking University)

 

In this paper we propose a new concept of measuring heteroskedasticity based on nonparametric quantile regression. This concept is also used to investigate the impact of heteroskedasticity on the efficiency/performance of the tests for the significance of the coefficients of the variables of interest.

 

A Simultaneous Test for Model Specification of Nonlinear Time Series Models (Shuo Li, Bin Guo and Yundong Tu)

Speaker: Shuo Li (Tianjin University of Finance and Economics)

 

This paper proposes a simultaneous test for the specification of the conditional mean and conditional variance functions as well as the error distribution in nonlinear time series models. Constructed by comparing different density estimators for the response variable, the proposed test has a Gumbel limiting distribution under the null hypothesis and is consistent against the alternative hypothesis. A simulation-based scheme is proposed to calibrate the asymptotic null distribution. The proposed test is shown to have nice finite sample property in simulations. The application to the continuous time diffusion model is illustrated via an analysis on the U.S. Federal fund rate data.

 

Section II:  Bayesian Econometrics

 

Bayesian Specification Testing for Latent Variable Models (Yong Li, Tao Zeng and Jun Yu)

Speaker: Tao Zeng (Wuhan University)

 

A Bayesian test statistic is proposed to assess the model specification after the model is estimated by Bayesian MCMC methods. The proposed approach does not require an alternative model to be specified and is applicable to a variety of models, including latent variable models for which frequentist methods are difficult to use. The properties of the test statistic are established and its implementation is discussed. The test is easy to use and the test statistic can be calculated from MCMC outputs even when there are latent variables. The finite sample properties are investigated using simulated data. The method is illustrated in a linear regression model, a dynamic factor model, and a stochastic volatility model using real data

 

Functional Coefficient Time Series Models with Trending Regressors

Speaker: Tingting Cheng (Nankai University)

 

This paper studies a functional coefficient time series model with trending regressors, where the coefficients are unknown functions of time and random variables. We propose a local linear estimation method to estimate the unknown coefficient functions. An asymptotic distribution of the proposed local linear estimator is established under mild conditions. A test procedure is developed to test the null hypothesis that the functional coefficients take particular parametric forms. For practical use, we further propose a Bayesian approach to select bandwidths involved in this local linear estimator. Several numerical examples are provided to examine the finite sample performance of the proposed local linear estimator and the test procedure. The results show that the local linear estimator works well and the proposed test has satisfactory size and power. In addition, simulation studies show that the Bayesian bandwidth selection method is better than cross–validation method. Furthermore, we employ the functional coefficient model to study the relationship between consumption per capita and income per capita in U.S. and the results show that functional coefficient model with our proposed local linear estimator and Bayesian bandwidth selection method performs best in both in–sample fitting and out–of–sample forecasting.

 

A New Bayesian Estimator for Life-Cycle Model

Speaker: Xiaobing Liu (Singapore Management University)

 

In this paper, we develop a new Bayesian method for estimating finite-horizon dynamic life-cycle model with continuous state variables. The estimation method avoids the drawbacks of the method based on the log-linearized Euler equation. It does not require the error distribution to be specified. It is easy to compute. When the optimal policy function is available analytically, we propose a simple and fast algorithm to compute the estimator, which is easy to parallelize. In this case, we also find the nice asymptotic properties of our estimator. In particular, the estimator is asymptotically normal with the mean being the true value of the parameter. In addition we derive the over-identification and point-null hypothesis tests based on the Bayesian large sample theory. When a numerical method is used to approximate the optimal policy function, we show that, under mild conditions, our estimate continues to enjoy the nice asymptotic properties as if the optimal policy function is available analytically.

 

Section IIILarge Panel Data Models

 

Quantile Factor Models (Liang Chen, Juan J. Dolado, and Jesús Gonzalo)

Speaker: Liang Chen (Shanghai University of Finance and Economics)

 

In this paper we introduce a novel concept: Quantile Factor Models (QFM), where a few unobserved common factors may affect all parts of the distributions of many observed variables in a panel dataset of dimension N × T. When the factors affecting the quantiles also affect the means of the observed variables, a simple two-step procedure is proposed to estimate the common factors and the quantile factor loadings. Conditions on N and T ensuring uniform consistency and weak convergence of the entire quantile factor loadings processes differ from standard conditions in factor-augmented regressions with smooth object functions. Based on these results, we show how to make inference on the quantile factor loadings in a location-scale shift factor model. When factors affecting the quantiles differ from those affecting the means of the observed variables, we propose an iterative procedure to estimate both factors and factor loadings at a given quantile. Simulation results confirm a satisfactory performance of our estimators in small to moderate sample sizes. In particular, it is shown that the iterative procedure can consistently estimate common factors that cannot be captured by PC estimators. Empirical applications of our methods to several datasets of financial returns are considered.

 

Specification Tests for Large Dynamic Panel Data Models (Yoonseok Lee and Bin Peng)

Speaker: Bin Peng (Huazhong University of Science and Technology)

 

This paper considers the tests of specification, including the tests for serial correlation and the tests of overidentifying restrictions, for large dynamic panel data models. The test statistics are built upon the two-step GMM estimations using three different instrument matrices: the block-diagonal matrix with a full set of all available instruments, the block-diagonal matrix with a subset of all available instruments and the collapsed matrix with a subset of all available instruments. It shows that the conventional Sargan`s test of overidentifying restrictions (Arellano and Bond (1991)) does not approximate to the chi-square distribution, when the number of instruments used, is relatively as large as N; therefore it proposes corrected Sargan`s tests with different instrument matrices. The limiting distributions of all the tests of specification are derived as N and T go to infinity simultaneously. Power properties are discussed under varieties of alternatives. The results show that the tests for serial correlation are powerful against different alternatives, and the power of corrected Sargan`s tests only increases as N increases. Monte Carlo simulations confirm our theoretical findings, especially showing that the corrected Sargan`s tests have the correct size. Besides, it suggests using the collapsed instrument matrix for the practical testing purpose.

 

Dynamic Factor VaR Measurement for Large Portfolios in an Emerging Market (Yanyan Guo and Yanyun Zhao)

Speaker: Yanyun Zhao (Zhongnan University of Economics and Law)

 

This paper proposes a new approach that combines dynamic factor model (DFM) and asymmetric generalized dynamic conditional correlation (AG-DCC) model to detect time-varying variances and correlations for a large number of financial variables and then measure the Value-at-Risk (VaR). By extracting the information of large portfolios on a small number of factors and permitting asymmetry in them, this methodology relieves computation burden and advances prediction accuracy simultaneously, which is shown to be more efficient than the widely used VaR methodologies. The empirical analysis has been carried out in a large portfolio from Chinese stock market.

 

Section IV:  Econometric Applications

 

Forecasting with Social Media Data in the Presence of Heteroscedasticity (Steven Lehrer and Tian Xie)

Speaker: Tian Xie (Xiamen University)

 

The big data market has been growing at an exceptional pace and the introduction of more sophisticated strategies to conduct forecasts has been driven by the machine learning literature. Many of these developments aim to deal with heteroscedastic data, a feature well studied by econometricians. In this note, we conduct a set of simulation experiments contrasting both LASSO-based and non-LASSO strategies using data from the _lm industry in conjunction with social media. We find that there are large efficiency gains from using model averaging estimators suggesting that econometric strategies that deal with model uncertainty over practical benefits for practitioners and researchers conducting forecasts with new big data sources.

 

Chinese Trade Price and Yuan`s Valuation (Yichen Gao, Li Gan, and Qi Li)

Speaker: Yichen Gao (Capital University of Economics and Business)

 

This paper studies how the major economic events would have affected Chinese Yuan`s nominal exchange rate against US dollar from 1989 to 2013, had China adopted a more flexible exchange rate policy. These events include the return of Macao to China in December 1999, China`s accession to World Trade Organization (WTO) in December 2001, and the reform of Chinese exchange rate policy (July 2005 and June 2010). During 1994-2005, Chinese Yuan was pegged to US dollar, therefore, the traditional average treatment effect estimation methods cannot be used to consistently estimate Yuan`s exchange rate under the counterfactual scenario of a free oat exchange rate policy. We develop a new estimation strategy by combining a novel panel data method (proposed by Hisao, Ching and Wan 2012) and the purchasing power parity theory. Based on this new estimation strategy, we find that during the period of pegging to US dollar, Chinese Yuan were undervalued. Due to the return of Macao, Chinese Yuan was undervalued by about 15.96%; China`s accession to WTO caused Chinese Yuan undervalued even more by 36.60%. After the reform of Chinese exchange rate policy in July 2005, the undervaluation of Yuan was reduced to only 0.76%; the policy reform in June 2010 made Yuan overvalued by 14.40%. We conclude that Chinese fixed exchange rate policy from 1994-2005 indeed undervalued Yuan, especially after its accession to WTO. The reform of Chinese exchange rate policy were making progress in re-evaluating Yuan.

 

The Impact of Export Expansion on Youth’s Education: Triple Difference Evidence from China’s WTO Accession

Speaker: Faqin Lin (Central University of Finance and Economics)

 

This paper presents empirical evidence that the growth of export manufacturing in China after its WTO accession in 2001 altered the youth’s education decision. I exploit variation in exposure to WTO across cohorts, regions and sectors to identify the effect of export expansion on education by triple difference regression techniques. I document statistically significant decreases in Chinese high school dropout rates for youths most affected by expansions in export manufacturing. The magnitudes I find suggest that for every 100 youth finishing junior middle school, export expansion causes 3.5 to 7 students to drop out rather than continuing through to high middle school. I further show that export expansion contributes to the loss of education by offering less-skilled processing exporting jobs which raised the opportunity cost of schooling for students at the margin. Such reduction is mainly driven by the young from rural areas and from families with siblings.

 

Section V: Financial Econometrics

 

Adaptive Estimation of Functional-coefficient Cointegration Models with Nonstationary Volatility

(Yundong Tu and Ying Wang)

Speaker: Ying Wang (Peking University)

 

This paper analyzes functional-coefficient cointegration models with nonstationary (unconditional) volatility of a general form. The kernel weighted least squares (KLS) estimator of Xiao (2009) is subject to potential efficiency loss, and can be improved by an adaptive kernel weighted least squares (AKLS) estimator that adapts to heteroscedasticity of unknown form. The AKLS estimator is shown to be as efficient as the generalized kernel weighted least squares estimator asymptotically, and can achieve significant efficiency gain relative to the KLS estimator infinite samples. An illustrative example is provided by investigating the purchasing power parity hypothesis.

 

 

Portfolio Analysis of Skewed Data with Risk Measure (Ruili Sun Tiefeng Ma)

Speaker: Ruili Sun (Southwestern University of Finance and Economics)

 

The covariance matrix that measures the risk and relationship of assets returns simultaneously plays a crucial role in the minimum variance portfolio framework. Therefore, based on the advantages of semi-variance and distance correlation, a new risk measurement tool is proposed by using the decomposition of covariance matrix in this paper. The covariance matrix can be decomposed as the product of the diagonal matrix whose elements are assets returns variances and the matrix of the correlation coefficients of assets returns, which the variances measures the risk and correlation coefficients measures the relationship between assets returns. In order to more fully measure the risk and relationship of assets returns, a new risk measurement tool based on the decomposition of covariance matrix is constructed by using semi-variance and distance correlation. The new risk measurement tool is constructed by rewriting the counterparts of the decomposition of covariance matrix as follows: (i) using the semi-variance to substitute for the variance, (ii) using the distance correlation to replace the Pearson correlation. The general covariance matrix is replaced by the new risk measurement tool. The proposed portfolio optimization strategy is applied to empirical data and the numerical studies show a good performance of this strategy.

 

Conditional Covariance Model

Speaker: Jin Liu (Peking University)

 

Estimation of the conditional covariance matrices is an important topic in statistics and finance. The pure nonparametric estimator and its asymptotic properties had been studied by Yin et al. (2010). To overcome “curse of dimensionality” in Yin`s nonparametric estimation, we propose a averaging estimator for the conditional covariance matrices, which combines the method of MAMAR (Model Averaging MArginal Regression) by Li Linton and Lu (2015), and covariance regression by Zou et al. (2016). The asymptotic properties are derived to justify the estimation procedure and simulation studies illustrate its performance in finite sample. An empirical application in portfolio allocation is used to show the usefulness of proposed estimator.