Panel Threshold Regression with Unobserved Individual-Specific Threshold Effects
Grant Data
Project Title
Panel Threshold Regression with Unobserved Individual-Specific Threshold Effects
Principal Investigator
Dr Yu, Ping
(Principal Investigator (PI))
Duration
30
Start Date
2019-09-01
Amount
450000
Conference Title
Panel Threshold Regression with Unobserved Individual-Specific Threshold Effects
Presentation Title
Keywords
Individual-Specific Effect, Panel Data, Specification Testing, Threshold Regression, Wild Bootstrap
Discipline
Econometrics and Business Statistical Methods
Panel
Business Studies (B)
HKU Project Code
17500719
Grant Type
General Research Fund (GRF)
Funding Year
2019
Status
Completed
Objectives
1 We will develop a correlated random effects estimation method when the individual-specific effects are regime-dependent. 2 Specification tests will be constructed to check the sufficiency of the assumed correlated random effects framework to control endogeneity and to determine whether the individual-specific effects are indeed regime-dependent. 3 Our estimation and testing methods will be extended to handle more general patterns of endogeneity in panel data. 4 Simulation studies will be conducted to check the validity and finite sample performance of our estimation methods and tests. 5 The methods will be applied to a real data set to check their performance and assess the importance of endogenous and regime dependent thresholding in practical work.