Heterogeneous causal inference of corporate profitability using two-scale distributional nearest neighbor estimation

HE Yu; WU Yue; ZHENG Zemin; WU Jie

doi:10.3969/j.issn.0253-2778.2020.12.006

PDF( 1867 KB)

Open Access JUSTC Research Article

Heterogeneous causal inference of corporate profitability using two-scale distributional nearest neighbor estimation

1.
School of Management, University of Science and Technology of China, Hefei 230026
2.
School of Insurance, Central University of Finance and Economics, Beijing 100081

Funds: This work was supported by National Natural Science Foundation of China (72071187, 11901603, 11671374, 71731010, and 71921001), and Fundamental Research Funds for the Central Universities (WK347000 0017 and WK2040000027).

Cite this:

https://doi.org/10.3969/j.issn.0253-2778.2020.12.006

More Information

Author Bio:
He Yu and Wu Yue are co-first authors. He Yu is a graduate student under the supervision of Prof. Zheng Zemin at University of Science and Technology of China (USTC). Her current research is mainly focused on statistical inference and applications. Wu Yue received her Ph. D. degree from State University of New York. She is currently an associate professor at Central University of Finance and Economics. Her research interests cover risk management, dynamic optimization and insurance.
Corresponding author: Zheng Zemin (corresponding author) is Full Professor at the department of management, University of Science and Technology of China. He received his B. S. degree from USTC in 2010 and Ph. D. degree from University of Southern California in 2015. Afterwards, he was working in Department of Management, University of Science and Technology of China. His research interests include high dimensional statistical inference and big data problems.; Wu Jie (corresponding author) is currently a Ph. D. student from School of Management, University of Science and Technology of China. She received her B. S. degree from Anhui University of Technology in 2017. Her research mainly focuses on high dimensional variable selection, classification.
Received Date: 09 September 2020
Accepted Date: 17 December 2020
Publish Date: 30 December 2020

Abstract Full text PDF

Abstract

Abstract

The heterogeneous treatment effect is the center of gravity in modern causal inference and is also widely used in the financial field. Studying the heterogeneous effect of improving solvency on the profitability of listed companies with different operational capabilities can eliminate the impact of individual differences in listed companies and make the estimation results more accurate. The two-scale distributional nearest neighbor (DNN) is able to eliminate the first-order finite sample bias in the estimate, and it is asymptotically unbiased and asymptotically normal under regularization conditions. Therefore, an empirical analysis was conducted of the financial indicators of Chinese listed companies in the CSMAR financial statement database by using the two-scale distributional nearest neighbor estimation. And a study was carried out on that whether there is heterogeneity in the impact of corporate solvency reduction on corporate profitability under different operating capabilities. Corresponding suggestions were given for optimizing the financial structure and improving the profitability of listed companies.

Abstract

The heterogeneous treatment effect is the center of gravity in modern causal inference and is also widely used in the financial field. Studying the heterogeneous effect of improving solvency on the profitability of listed companies with different operational capabilities can eliminate the impact of individual differences in listed companies and make the estimation results more accurate. The two-scale distributional nearest neighbor (DNN) is able to eliminate the first-order finite sample bias in the estimate, and it is asymptotically unbiased and asymptotically normal under regularization conditions. Therefore, an empirical analysis was conducted of the financial indicators of Chinese listed companies in the CSMAR financial statement database by using the two-scale distributional nearest neighbor estimation. And a study was carried out on that whether there is heterogeneity in the impact of corporate solvency reduction on corporate profitability under different operating capabilities. Corresponding suggestions were given for optimizing the financial structure and improving the profitability of listed companies.