Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/142092
Authors: 
Ajibola, Isaiah O.
Udoette, Ubong S.
Omotosho, Babatunde S.
Muhammad, Rabia A.
Year of Publication: 
2015
Citation: 
[Journal:] CBN Journal of Applied Statistics [ISSN:] 2476-8472 [Publisher:] The Central Bank of Nigeria [Place:] Abuja [Volume:] 6 [Year:] 2015 [Issue:] 1 [Pages:] 111-132
Abstract: 
This study investigates the long run relationship between exchange rate andexternal reserves in Nigeria during 1990Q1 - 2012Q4. We confirm theexistence of threshold cointegration between the variables in Nigeria, asagainst linear cointegration. Consequently, a two-regime threshold vector error correction model (TVECM) is estimated via maximum likelihood procedure. Model results indicate that cointegration between the variables occurs only when the equilibrium error exceeds an estimated threshold parameter of 0.52. Having partitioned the TVECM into two regimes based on the obtained threshold, we find that the error correction coefficients of the exchange rate in the two regimes are not significant, implying that exchange rates do not respond to equilibrium error during the estimation period. On the other hand, external reserves adjust to correct past divergence, albeit only when the equilibrium error exceeds the threshold parameter. Overall, external reserves adjust to maintain long run equilibrium while exchange rates do not, which seems to align with the monetary authority's action of deploying external reserves to maintain exchange rate stability in the country.
Subjects: 
Exchange rates
External reserves
Threshold cointegration
Threshold error correction
JEL: 
C22
F31
Document Type: 
Article

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.