Research Interests
Applied Microeconometrics, Financial Econometrics, and Theoretical Econometrics.
Research Papers
“Direct Nonparametric Conditional Quantile Estimation for Time Series Data” (Job Market Paper)
In this paper we consider the problem of forecasting a conditional quantile function in a nonparametric framework with time series data. We prove the consistency and asymptotic normality of our nonparametric conditional quantile estimator for absolutely regular processes (β -mixing) data generating processes. We derive the asymptotic distribution of our proposed estimator and conduct Monte Carlo experiments to compare the finite sample performance of our estimator and traditional check function based estimator. Simulation results show that our estimator outperforms the check function based estimator in terms of out-of-sample forecasting. We also apply the estimation method to forecast the monthly U.S. housing return based on S&P/Case-Shiller House Price Indices and compare the forecasting results with those obtained by using a commonly used linear conditional quantile model and by using conventional check function based nonparametric conditional quantile estimator. Our estimator forecast well compared to these competitors, especially for data in the tail regions.
“An Alternative Bandwidth Selection Method for Estimating Functional Coefficient Models”, submitted to Economics Letters (with Ta-Cheng Huang and Qi Li).
Functional coefficient regression models are very useful for many statistics and economics applications and there exists a large body of literature on kernel estimation of the coefficient functions. Fan and Zhang (1999) point out that the traditional estimation method is not optimal if the coefficient functions possess different degrees of smoothness. They propose a two-step method to attenuate the drawback. To apply the two-step method, one needs to identify which coefficient function is the smoothest one and undersmooth all other functions in the first stage. In this paper we propose an alternative approach which assigns each smooth function a different bandwidth and we choose the bandwidths simultaneously. Simulation results show that our proposed method compare well with existing methods.
“Nonparametric Identification and Estimation of Additive Social Interaction Models with Homophily,” (with Wenzheng Gao and Darong Dai)
This paper studies strategic social interaction among economic agent that is connected through the effect of homophily. In particular, we measure the homophily effects by adopting a social network distance function that is monotonically decreasing in the difference between players' socioeconomic characteristics. Furthermore, under a symmetric equilibrium selection mechanism, we establish a nonparametric approach to identify the structural model and propose a computationally feasible two-step estimation procedure. The asymptotic properties of the two-step estimator are derived under context of ``large games", i.e., the number of players going to infinity. Finally, we apply the identification and estimation methods to study the peer effects on youth smoking behaviors using data of adolescents in the United States and find positive and statistically significant peer effects and demonstrate the empirical importance of including homophily effects in our model.
“Identifying Housing Vacancy Rate from Electricity Bills,” (with Li Gan and Qi Li).
We investigate the reasons for the rapid price growth of Chinese residential homes through vacancy rate channel. We found out the vacancy rate of 15.85\% from household-level electricity consumption bills data set by mixture density model. We conclude that the strong investment demands contributing to the rapid price growth and resulting in the over-supply condition of Chinese housing market. These investment properties could hit the market and put more pressure on downward price when housing price decrease since the supply of vacant houses are more elastic than that of occupied ones. We suggest that levying property tax could be helpful in suppressing investment demand among private households.
Retailer Strategic Pricing under Sticky Demand, (with Haipeng Chen, Haoying Sun and Zheng Li).
The literature on dynamic pricing assumes that consumer demand responds to any changes in price. Recent advances in economics, however, have suggested that consumers may be rationally inattentive and not respond to small price changes, resulting in demand stickiness. We explicitly model the implications of this sticky demand on firm’s pricing behaviors. Using a large dataset consisting of eight years of weekly grocery retail data, we estimate the magnitude of demand stickiness and demonstrate how the estimates vary with consumer demographics and product characteristics. Furthermore, we conduct a counterfactual analysis demonstrating the profit improvement a retailer could enjoy by taking into account this demand stickiness when dynamically setting prices to clear inventory.
Work in Progress
“Can Incentives Hurt: Evidence from the Ride-Sharing Market”, (with Zheng Li, Liu Ming, and Weiming Zhu).
A new study of driver’s labor supply decisions under ride sharing context (with Zheng Li, Liu Ming, Weiming Zhu). With a comprehensive dataset obtained from Didi, our paper aims to resolve two following questions. First, do Didi Kuaiche drivers exhibit reference-dependent preferences for labor supply decisions? Second, given their reference-dependent preferences, what is the optimal bonus scheme to maximize firm's revenue? To this end, we first offer empirical evidence that Didi driver are reference dependent. Then, we model driver's utility with different reference points and structurally estimate the corresponding reference points.
“Identity Discrimination, Market Structure and Wage Inequality”, (with Darong Dai and Guoqiang Tian).
A study of how the change of market structure may affect equilibrium wage inequality induced by identity discrimination based on multiple survey datasets from China (with Darong Dai and Guoqiang Tian). For both labor market and capital market, we study two market structures: price taking and bilateral bargaining. We hence compare four market-structure combinations. We discuss between-group wage inequality using three different while related indexes: wage-rate inequality, wage-income inequality and aggregated wage-income inequality. We provide a full characterization on the change of wage inequality across alternative market-structure combinations. In particular, we show how production technology and preference structure may change the effect of market structure transformation placed on wage inequality.
“Investment and Consumption Demand in the Chinese Housing Market”, (with Li Gan).
A study of the high multiple homes ownership rate and vacancy rate with faster than economic fundamentals housing price growth by disentangling the consumption and investment housing demand from China Household Finance Survey data (with Li Gan).
Applied Microeconometrics, Financial Econometrics, and Theoretical Econometrics.
Research Papers
“Direct Nonparametric Conditional Quantile Estimation for Time Series Data” (Job Market Paper)
In this paper we consider the problem of forecasting a conditional quantile function in a nonparametric framework with time series data. We prove the consistency and asymptotic normality of our nonparametric conditional quantile estimator for absolutely regular processes (β -mixing) data generating processes. We derive the asymptotic distribution of our proposed estimator and conduct Monte Carlo experiments to compare the finite sample performance of our estimator and traditional check function based estimator. Simulation results show that our estimator outperforms the check function based estimator in terms of out-of-sample forecasting. We also apply the estimation method to forecast the monthly U.S. housing return based on S&P/Case-Shiller House Price Indices and compare the forecasting results with those obtained by using a commonly used linear conditional quantile model and by using conventional check function based nonparametric conditional quantile estimator. Our estimator forecast well compared to these competitors, especially for data in the tail regions.
“An Alternative Bandwidth Selection Method for Estimating Functional Coefficient Models”, submitted to Economics Letters (with Ta-Cheng Huang and Qi Li).
Functional coefficient regression models are very useful for many statistics and economics applications and there exists a large body of literature on kernel estimation of the coefficient functions. Fan and Zhang (1999) point out that the traditional estimation method is not optimal if the coefficient functions possess different degrees of smoothness. They propose a two-step method to attenuate the drawback. To apply the two-step method, one needs to identify which coefficient function is the smoothest one and undersmooth all other functions in the first stage. In this paper we propose an alternative approach which assigns each smooth function a different bandwidth and we choose the bandwidths simultaneously. Simulation results show that our proposed method compare well with existing methods.
“Nonparametric Identification and Estimation of Additive Social Interaction Models with Homophily,” (with Wenzheng Gao and Darong Dai)
This paper studies strategic social interaction among economic agent that is connected through the effect of homophily. In particular, we measure the homophily effects by adopting a social network distance function that is monotonically decreasing in the difference between players' socioeconomic characteristics. Furthermore, under a symmetric equilibrium selection mechanism, we establish a nonparametric approach to identify the structural model and propose a computationally feasible two-step estimation procedure. The asymptotic properties of the two-step estimator are derived under context of ``large games", i.e., the number of players going to infinity. Finally, we apply the identification and estimation methods to study the peer effects on youth smoking behaviors using data of adolescents in the United States and find positive and statistically significant peer effects and demonstrate the empirical importance of including homophily effects in our model.
“Identifying Housing Vacancy Rate from Electricity Bills,” (with Li Gan and Qi Li).
We investigate the reasons for the rapid price growth of Chinese residential homes through vacancy rate channel. We found out the vacancy rate of 15.85\% from household-level electricity consumption bills data set by mixture density model. We conclude that the strong investment demands contributing to the rapid price growth and resulting in the over-supply condition of Chinese housing market. These investment properties could hit the market and put more pressure on downward price when housing price decrease since the supply of vacant houses are more elastic than that of occupied ones. We suggest that levying property tax could be helpful in suppressing investment demand among private households.
Retailer Strategic Pricing under Sticky Demand, (with Haipeng Chen, Haoying Sun and Zheng Li).
The literature on dynamic pricing assumes that consumer demand responds to any changes in price. Recent advances in economics, however, have suggested that consumers may be rationally inattentive and not respond to small price changes, resulting in demand stickiness. We explicitly model the implications of this sticky demand on firm’s pricing behaviors. Using a large dataset consisting of eight years of weekly grocery retail data, we estimate the magnitude of demand stickiness and demonstrate how the estimates vary with consumer demographics and product characteristics. Furthermore, we conduct a counterfactual analysis demonstrating the profit improvement a retailer could enjoy by taking into account this demand stickiness when dynamically setting prices to clear inventory.
Work in Progress
“Can Incentives Hurt: Evidence from the Ride-Sharing Market”, (with Zheng Li, Liu Ming, and Weiming Zhu).
A new study of driver’s labor supply decisions under ride sharing context (with Zheng Li, Liu Ming, Weiming Zhu). With a comprehensive dataset obtained from Didi, our paper aims to resolve two following questions. First, do Didi Kuaiche drivers exhibit reference-dependent preferences for labor supply decisions? Second, given their reference-dependent preferences, what is the optimal bonus scheme to maximize firm's revenue? To this end, we first offer empirical evidence that Didi driver are reference dependent. Then, we model driver's utility with different reference points and structurally estimate the corresponding reference points.
“Identity Discrimination, Market Structure and Wage Inequality”, (with Darong Dai and Guoqiang Tian).
A study of how the change of market structure may affect equilibrium wage inequality induced by identity discrimination based on multiple survey datasets from China (with Darong Dai and Guoqiang Tian). For both labor market and capital market, we study two market structures: price taking and bilateral bargaining. We hence compare four market-structure combinations. We discuss between-group wage inequality using three different while related indexes: wage-rate inequality, wage-income inequality and aggregated wage-income inequality. We provide a full characterization on the change of wage inequality across alternative market-structure combinations. In particular, we show how production technology and preference structure may change the effect of market structure transformation placed on wage inequality.
“Investment and Consumption Demand in the Chinese Housing Market”, (with Li Gan).
A study of the high multiple homes ownership rate and vacancy rate with faster than economic fundamentals housing price growth by disentangling the consumption and investment housing demand from China Household Finance Survey data (with Li Gan).