Dear researchers/econometricians/statisticians,
I have some confusion about deciding the stationarity testing for panel data.
My dataset comprises of 850+ companies (cross-sections) and 2001-2020 years (20 years unbalanced).
According to my understanding Panel, stationarity testing is only applicable when T>N but in my case N>T.
Some of the researchers say that the average number of the years should be greater than 10 to apply panel stationarity test (without proper reference and justification) while in my case these are 19 (unbalanced as for some year data was not available). So I got confused at this point.
Kindly do some confirmation along with references that what to do in my case? Whether I should apply stationarity test or not. If yes then please share some references that support stationarity testing for panel data involving N>T.
I made a quick search and found:
N Bennala, M Hallin, D Paindaveine (2012) Pseudo-Gaussian and rank-based optimal tests for random individual effects in large n small T panels, Journal of Econometrics 170 (1), 50-67
N Bennala, M Hallin, D Paindaveine (2010) Rank-Based Optimal Tests for Random Effects in Panel Data
ECARES Working Papers
and a thesis by Nazir Bennala at Université libre de Bruxelles, Belgium.
See these links:
https://www.sciencedirect.com/topics/economics-econometrics-and-finance/panel-unit-root
https://www.ukessays.com/essays/psychology/why-to-use-panel-unit-root-tests-psychology-essay.php
https://www.econstor.eu/bitstream/10419/19627/1/200542dkp.pdf
https://www.jstor.org/stable/25146517
Dear Zahid,
Please check:
1. Arellano, M & Bover, O (1995) in Journal of Econometrics 68, 29-52
2. Blundell, R & Bond, S (1998) in Journal of Econometrics 87(1), 115-143.
3. Roodman , D (2006) in Centre fir Global Development working paper 103, 1-45
Dear Kenneth Anyalechi
Thank you very much. I am grateful for this kindness.
- Badges
- Science topic
- Similar topics
- Philosophy
- Esthetics
- Interpretation