Big Data in Economics and Management
Description
With the rapid development of big data, three major challenges arise in the field of economics and management. The first challenge is that the traditional correlation-based methods cannot essentially reveal the true philosophy under the economic activities, modelling and inferring the causal relationship is paramount for discovering the essential effect of certain economic and management policies. The second one is that the computational burden becomes extremely high and the estimation accuracy is lost when the data scale is large. The third one is that financial institutions typically hold tens of thousands of assets, making portfolio risk assessment very computationally intensive.
This book discusses three advanced topics in modern economics and management: causal inference, financial model computing and decisions, and financial risk management. The first part of the book introduces the counterfactual framework for causal inference in observational studies and defines important causal parameters under both discrete and continuous treatments. The second part focuses on the computations associated with the financial model and its consequent decision making. The third part studies the nested simulation method for portfolio risk measurement and introduces the neural network methodology for market risk forecasting.
The goal of this book is to provide cutting-edge methodologies and rigorous theory to solve advanced problems in economics and management, such as program/policy evaluation, efficient computation of econometric models, and financial risk management. This book will be appealing to academic researchers and graduate students. Practitioners may also find this book helpful.
This is an open access book.
More details
Persons
Zheng Zhang is an Associate Professor (with tenure) at the Institute of Statistics and Big Data, Renmin University of China. He earned a Ph.D. in 2015 from the Department of Statistics, Chinese University of Hong Kong. His research fields mainly focus on causal inference and econometrics.
Kun Zhang is an Assistant Professor at the Institute of Statistics and Big Data, Renmin University of China. He obtained a bachelor degree in Mathematics and Applied Mathematics, master degree in Probability Theory from Beijing Normal University, and doctoral degree in Management Science from City University of Hong Kong. His research interests include stochastic simulation, machine learning, financial engineering and risk management.
Xing Yan is an Assistant Professor at the Institute of Statistics and Big Data, Renmin University of China. He obtained his Ph.D. degree from the Chinese University of Hong Kong in 2019. He works at the intersection of AI and finance. He does research on problems in the areas of financial engineering and FinTech such as risk management, asset pricing, quantitative investments, and derivatives, with machine learning and data science methodologies. His research interests also include generative learning, causal learning, OOD generalization, and uncertainty quantification in machine learning.
Songshan Yang is an Assistant Professor at the Institute of Statistics and Big Data, Renmin University of China. He obtained a bachelor degree in Statistics from Beijing Normal University, and doctoral degree in Statistics from Pennsylvania State University. His research interests include high dimensional data analysis, statistical optimization, machine learning and applications of statistical models in finance, physiology and psychology.
Yuqian Zhang is an Assistant Professor at the Institute of Statistics and Big Data, Renmin University of China. He obtained his Ph.D.from the University of California San Diego in 2022 and a bachelor's degree from Wuhan University in 2016. His research focuses on theory and methodology in causal inference, missing data problems, semi-supervised inference, high-dimensional statistics, and machine learning.
Content
Part I Causal Inference in Economics.- 1 Causal Inference for A Discrete Treatment.- 1.1 Basic Framework.- 1.2 Causal Inference based on Covariate Balancing Calibration.- 1.3 Causal Inference based on Semi-supervised Data.- 1.4 Causal Inference based on Neural Networks .- 2 Causal Inference for A Continuous Treatment.- 2.1 Basic Framework.- 2.2 Semiparametric Efficiency Bound.- 2.3 Maximum Entropy Weighting.- 2.4 Efficient Estimation Results.- 2.5 Model Specification Tests.- 2.6 Nonparametric Estimation of ATE .- 2.7 Nonparametric Estimation of Distributional and Quantile Treatment Effects.- 2.8 Testing Distributional Effects.- 2.9 Empirical Study: Presidential Campaign Data.- 3 Causal Inference with Measurement Errors.- 3.1 Basic Framework.- 3.2 Estimation Method.- 3.3 Large Sample Properties.- 3.4 Select the Smoothing Parameters.- 3.5 Real Data Example.- Part II Financial Model Computing and Decisions.- 4 Efficient Computing for High-Dimensional Econometric Models.- 4.1 Introduction.- 4.2 Asset-splitting algorithm for portfolio selection.- 4.3 Feature-splitting algorithm for PQR.- 4.4 Numerical study.- 4.5 Conclusion and discussion .- 4.6 Appendix.- Part III Financial Risk Management.- 5 Bootstrap-based Budget Allocation for Nested Simulation.- 5.1 Introduction.- 5.2 Backgrounds.- 5.3 A Sample-Driven Budget Allocation Method .- 5.4 Appendix .- 6 Constructing Confidence Intervals for Nested Simulation.- 6.1 Introduction.- 6.2 Formulations .- 6.3 Confidence Intervals .- 7 Deep Probabilistic Forecasting for Market Risks.- 7.1 Background of Market Risk Forecasting .- 7.2 Background of Uncertainty Quantification in Machine Learning .- 7.3 Deep Sequential Learning of Conditional Heavy-Tailed Distributions .- 7.4 Ensemble Multi-Quantile Regression with Deep Learning .- Appendix .- References .