Non-parametric Bayesian updating and windowing with kernel density and the kudzu algorithm

2022 ◽  
Vol 1 (1) ◽  
pp. 1
Author(s):  
Robert L. Grant
Keyword(s):  
Kernel Density ◽  
Bayesian Updating ◽  
Non Parametric

The dependency measures of commercial bank risks: Using an optimal copula selection method based on non-parametric kernel density

2020 ◽  
Vol 37 ◽  
pp. 101706
Author(s):  
Chenglu Jin ◽  
Rongda Chen ◽  
Diandian Cheng ◽  
Sitian Mo ◽  
Ke Yang
Keyword(s):  
Commercial Bank ◽  
Kernel Density ◽  
Selection Method ◽  
Bank Risks ◽  
Non Parametric

scholarly journals Explaining Disparities in AH Stock Premium Returns with Chinese Economic Policy Uncertainty: Evidence Based on a Non-Parametric Analysis

2019 ◽  
Vol 23 (4) ◽  
pp. 658-666 ◽  
Author(s):  
Mohib Ur Rahman ◽  
Bing Xu ◽  
Yuan Xiao ◽  
Wen-Tsung Lai ◽  
◽  
...  
Keyword(s):  
Economic Policy ◽  
Parameter Uncertainty ◽  
Parametric Analysis ◽  
Kernel Density ◽  
Evidence Based ◽  
Policy Uncertainty ◽  
Economic Policy Uncertainty ◽  
Premium Puzzle ◽  
The Impact ◽  
Non Parametric

This paper empirically investigates the AH share premium puzzle considering the impact of economic policy uncertainty (EPU) in China, using Hang Seng AH premium (HSAHP) index data from March 2011 to June 2018. Specifically, the index of Baker, Bloom, and Davis is used as a proxy for EPU in China. The data has been divided into two periods: 0 and 1. Period 0 includes data before the launch of the stock connect program and spans from March 2011 to October 2014, while period 1 represents data from after the launch of stock connect program and spans from November 2014 to June 2018. To more robustly test the change in AH premiums after the “connect” is launched; we evaluate the impact of Chinese EPU using non-parametric kernel density estimation. The empirical results show that parameter uncertainty explains variations in price disparity and can significantly reduce the returns of the AH share premium index.

Sign in / Sign up

Export Citation Format

Share Document