Estimation of Seasonal Variances in Descriptive Time Series Analysis
Kelechukwu C. N. Dozie *
Department of Statistics, Imo State University, Owerri, Imo State, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
This study presents the Buys-Ballot estimation procedure of row and seasonal variances in time series decomposition when trend component of time series is linear. Therefore, the main objective of this study is to obtain the Buys-Ballot estimates of row and seasonal variances for the mixed model. The method adopted in this study is Buys-Ballot procedure developed for choice of appropriate model for decomposition of any study series. The statistical software (MINITAB 17.0 version) is also adopted in this study. Results of the Buys-Ballot estimates for mixed model indicate that, the row variance is a function of trend parameters and seasonal component of the original series. It is for column variance, a constant multiple of the square of the seasonal component.
Keywords: Time series decomposition, trend-cycle component, mixed model, seasonal variance, Buys-Ballot estimate, choice of model