Please use this identifier to cite or link to this item:
Title: Long memory, seasonality and time trends in the average monthly rainfall in major cities of Nigeria
Authors: Yaya, O. S.
Gil-Alana, L. A.
Akomolafe, A. A.
Keywords: Long range dependence
Trend coefficient
Issue Date: Dec-2015
Publisher: The Central Bank of Nigeria
Abstract: Several features may be present in rainfall data, and sophisticated time series procedures are needed for the analysis. These features are that of seasonality, long range dependency of observations and time trend as observed in the climatological series. This paper therefore considered the analysis of these features in the monthly rainfall data of 37 meteorological stations across the six geo-political zones of Nigeria between 1981 and 2013. A fractional integration technique of time series analysis which permits the feature of rainfall time series to be examined in a single framework is employed. The procedure gives explicit formulas and directions that allow a moderately sophisticated analyst to perform trend analysis. The results show that the trend and persistence of long memory are fairly distributed across the six geopolitical zones such that a zone cannot be singled out with intense or abnormal rainfall distribution. By removing the seasonal variation and focusing on the rainfall anomalies with respect to the average monthly means, we found significant time trend coefficients that are positive, thus indicating that monthly rainfall has increased during the sampled periods.
ISSN: 2141-9272
Appears in Collections:Academic Publications in Statistics

Files in This Item:
File Description SizeFormat 
ui_art_yaya_long_2015.pdf578.33 kBAdobe PDFView/Open

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