I'M
I am a Ph.D. candidate in the Department of Economics, Cornell University. My research focuses on the intersection of econometrics and macroeconomics, with a particular emphasis on policy-related issues. I am particularly interested in using econometric techniques, such as partial identification, time series analysis, and Bayesian inference, to inform decision-making and improve policy outcomes. My work is both theoretical and applied, and I strive to bridge the gap between the two in my research.
My advisors are Francesca Molinari (chair), Yongmiao Hong, José Luis Montiel Olea, and Kristoffer Nimark.
I will be joining the Department of Economics at the University of Manchester as a Lecturer (Assistant Professor) this August.
WHAT I DO
Research
Working papers
Robust Bayesian Estimation and Inference for Dynamic Stochastic General Equilibrium Models （JMP）Partial Identification Bayesian
PDF
Abstract: This paper introduces a new algorithm to conduct robust Bayesian estimation and inference in dynamic stochastic general equilibrium models. The algorithm combines standard Bayesian methods with an equivalence characterization of model solutions. This algorithm allows researchers to perform the following analysis: First, find the complete range of posterior means of both the deep parameters and any parameters of interest robust to the choice of priors in a sense I make precise. Second, derive the robust Bayesian credible region for the model parameters. I prove the validity of this algorithm and apply this method to the models in Cochrane (2011) and An and Schorfheide (2007) to achieve robust estimations for structural parameters and impulse responses. In addition, I conduct a sensitivity analysis of optimal monetary policy rules with respect to the choice of priors and provide bounds to the optimal Taylor rule parameters.
Abstract: This paper introduces a new algorithm to conduct robust Bayesian estimation and inference in dynamic stochastic general equilibrium models. The algorithm combines standard Bayesian methods with an equivalence characterization of model solutions. This algorithm allows researchers to perform the following analysis: First, find the complete range of posterior means of both the deep parameters and any parameters of interest robust to the choice of priors in a sense I make precise. Second, derive the robust Bayesian credible region for the model parameters. I prove the validity of this algorithm and apply this method to the models in Cochrane (2011) and An and Schorfheide (2007) to achieve robust estimations for structural parameters and impulse responses. In addition, I conduct a sensitivity analysis of optimal monetary policy rules with respect to the choice of priors and provide bounds to the optimal Taylor rule parameters.
Centralized or Decentralized? An Empirical Model on Task Assignment of Government in Pandemics
(with Qiwei He) Partial Identification IO Time Series Bayesian
PDF Slides
Abstract: One of the central questions in pandemic-related discussions is how to best design mitigation policy to reduce the death rate from COVID-19. This paper enters into the debate by comparing social welfare using centralized and decentralized decision-making. Using indirect inference, we first estimate a structural SIR model with a regional spillover effect. We then set up and structurally estimate a dynamic game model where each US state government makes mitigation policy independently. Socially optimal mitigation policy is then solved by minimizing the sum of local governments' welfare loss using estimated weights on different sectors. Counterfactual analysis of centralized decision-making is conducted to compare the social welfare gain (loss) should the US adopt a mitigation policy at the federal level.
PDF Slides
Abstract: One of the central questions in pandemic-related discussions is how to best design mitigation policy to reduce the death rate from COVID-19. This paper enters into the debate by comparing social welfare using centralized and decentralized decision-making. Using indirect inference, we first estimate a structural SIR model with a regional spillover effect. We then set up and structurally estimate a dynamic game model where each US state government makes mitigation policy independently. Socially optimal mitigation policy is then solved by minimizing the sum of local governments' welfare loss using estimated weights on different sectors. Counterfactual analysis of centralized decision-making is conducted to compare the social welfare gain (loss) should the US adopt a mitigation policy at the federal level.
Nowcasting with dynamic factors: A LASSO penalized frequentist model averaging approach
(with Yongmiao Hong, Yuying Sun) Time Series ML
PDF Slides
Abstract: With the advent of complex information systems to collect data, real-time nowcasting faces various challenges, including a more significant number of predictors, higher order of lags, unbalanced data structure, model uncertainty and complexity in bridging high-frequency information contained. To address these issues, this paper proposes a new real-time nowcasting forecast combination with dynamic factor regressions, which deletes redundant predictors and simultaneously selects optimal weights for candidate models. We show that the selected weight achieves asymptotic optimality and consistency, even when all candidate models are misspecified. The proposed estimator is consistent and asymptotically Gaussian if the true model is included in candidate models. Simulation results show that the proposed method yields lower mean square forecast errors than alternative nowcasting methods, including MIDAS in Ghysels et al. (2004), GARS in Giannone et al. (2008), and FADL-MIDAS in Andreou et al. (2013). The proposed method is applied to forecast quarterly GDP with a set of 118 macroeconomic monthly data series, which compares favorably to other competing methods.
PDF Slides
Abstract: With the advent of complex information systems to collect data, real-time nowcasting faces various challenges, including a more significant number of predictors, higher order of lags, unbalanced data structure, model uncertainty and complexity in bridging high-frequency information contained. To address these issues, this paper proposes a new real-time nowcasting forecast combination with dynamic factor regressions, which deletes redundant predictors and simultaneously selects optimal weights for candidate models. We show that the selected weight achieves asymptotic optimality and consistency, even when all candidate models are misspecified. The proposed estimator is consistent and asymptotically Gaussian if the true model is included in candidate models. Simulation results show that the proposed method yields lower mean square forecast errors than alternative nowcasting methods, including MIDAS in Ghysels et al. (2004), GARS in Giannone et al. (2008), and FADL-MIDAS in Andreou et al. (2013). The proposed method is applied to forecast quarterly GDP with a set of 118 macroeconomic monthly data series, which compares favorably to other competing methods.
Work in progress
Venture Capital Investment Geography
(with Qinshu Xue, Bin Zhao) Search & Matching IO
PDF Slides
Abstract: This paper studies the geographic concentration of VC investment in the US. We discovered that both VC firms and VC-backed companies are highly clustered in the Bay-Boston-NY area. Using the detailed Uniform Commercial Code (UCC) filings that are self-reported by lenders to stake a claim to specific pieces of collateral, we track the transaction of capital across firms. We propose the vintage capital market density as an essential determinant of VC investment concentration. Since young firms can benefit from cheap vintage capital while old firms can exit with a high scrap value, VC investments are attracted due to a high entry rate and a low exit cost. By modeling a market for the vintage capital, we aim to endogenize the scrap value of firms. Our paper highlights the critical role of the capital market in determining the industry and venture capital agglomeration. Industry policy that helps promote local capital market density would also attract VC investment in places with the greatest economic need.
(with Qinshu Xue, Bin Zhao) Search & Matching IO
PDF Slides
Abstract: This paper studies the geographic concentration of VC investment in the US. We discovered that both VC firms and VC-backed companies are highly clustered in the Bay-Boston-NY area. Using the detailed Uniform Commercial Code (UCC) filings that are self-reported by lenders to stake a claim to specific pieces of collateral, we track the transaction of capital across firms. We propose the vintage capital market density as an essential determinant of VC investment concentration. Since young firms can benefit from cheap vintage capital while old firms can exit with a high scrap value, VC investments are attracted due to a high entry rate and a low exit cost. By modeling a market for the vintage capital, we aim to endogenize the scrap value of firms. Our paper highlights the critical role of the capital market in determining the industry and venture capital agglomeration. Industry policy that helps promote local capital market density would also attract VC investment in places with the greatest economic need.
Private and common information acquisition over the business cycle: Evidence from probability forecasts
(with Nathan Mislang, Kristoffer Nimark) Information Acquisition Bayesian
PDF Slides
Abstract: We propose a method to decompose a cross-section of observed belief revisions into private and common components. We define a common signal as the single signal that if observed by all agents can explain the maximum amount of belief revisions across agents. Private signals are defined to explain the remaining residual belief revision unaccounted for by the common signal. When applied to probability forecasts from the Survey of Professional Forecasters we find that on average, private information accounts for more of the observed belief revisions than common information. However, private information tend to increase uncertainty, i.e. to lead forecasters to assign higher probability to more extreme outcomes. The importance and precision of private and common information are positively correlated over time, though the importance and precision of private information is more volatile and more strongly correlated with business cycle indicators. We argue that no existing theoretical model can explain the documented facts. The proposed method is non-parametric and only assumes that agents use Bayes rule to update their beliefs.
PDF Slides
Abstract: We propose a method to decompose a cross-section of observed belief revisions into private and common components. We define a common signal as the single signal that if observed by all agents can explain the maximum amount of belief revisions across agents. Private signals are defined to explain the remaining residual belief revision unaccounted for by the common signal. When applied to probability forecasts from the Survey of Professional Forecasters we find that on average, private information accounts for more of the observed belief revisions than common information. However, private information tend to increase uncertainty, i.e. to lead forecasters to assign higher probability to more extreme outcomes. The importance and precision of private and common information are positively correlated over time, though the importance and precision of private information is more volatile and more strongly correlated with business cycle indicators. We argue that no existing theoretical model can explain the documented facts. The proposed method is non-parametric and only assumes that agents use Bayes rule to update their beliefs.
My Experience
Resume
Education
Ph.D. in Economics
2017-2023
Cornell University
M.S. in Economics
2015-2017
University of Wisconsin-Madison
B.S. in Economics
2011-2015
Southwest Jiaotong University.
Research Experience
Cornell University
Research Assistant
The Safety Net as a Springboard? A General Equilibrium based Policy Evaluation.
(Domenico Ferraro, Nir Jaimovich, Francesca Molinari and Cristobal Young, working paper)
Spring 2021
Cornell University
Research Assistant
Estimating the COVID-19 Infection Rate: Anatomy of an Inference Problem.
(Charles Manski and Francesca Molinari，Journal of Econometrics 2020)
Summer 2020
Cornell University
Research Assistant
Confidence Intervals for Projections of Partially Identified Parameters.
(Hiroaki Kaido, Francesca Molinari and Jorg Stoye, Econometrica 2019)
Spring 2020
TA
Teaching
Econ 6190: Econometrics I (Ph.D.)
Econ 6200: Econometrics II (Ph.D.)
Econ 1120: Introduction to Macroeconomics (Undergrad)
Econ 3120: Applied Econometrics (Undergrad)
Econ 1110: Introduction to Microeconomics (Undergrad)
Math 3110: Introduction to Analysis (Undergrad)
Math 2940: Linear Algebra for Engineers (Undergrad)
More about me
Contact
Let's connect
yk735@cornell.edu
Email
yizhou-kuang-982b16b5
Linkedin
kkkuang
Github
economics.cornell.edu
Cornell University