Premium
Common Breaks in Means for Cross‐Correlated Fixed‐ T Panel Data
Author(s) -
Westerlund Joakim
Publication year - 2019
Publication title -
journal of time series analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.576
H-Index - 54
eISSN - 1467-9892
pISSN - 0143-9782
DOI - 10.1111/jtsa.12407
Subject(s) - mathematics , estimator , consistency (knowledge bases) , panel data , monte carlo method , uncorrelated , cross section (physics) , statistics , algorithm , econometrics , discrete mathematics , physics , quantum mechanics
This note considers a panel data model in which the variable of interest has undergone a common structural break in the mean. The object of interest is the unknown breakpoint. The challenge is to device an estimator that is consistent when the data are cross‐correlated and the number of time periods T is fixed and cannot be increased without bound. The proposed solution involves taking an already existing estimator initially proposed for cross‐section uncorrelated panels and applying it to defactored data. Consistency is established as the number of cross‐section units N grows large, and is verified in small samples using Monte Carlo simulation.