versión On-line ISSN 2413-3051
versión impresa ISSN 1021-447X
J. energy South. Afr. vol.26 no.2 Cape Town may. 2015
In Volume 26 Number 1, the paper titled: A comparative study of the stochastic models and harmonically coupled stochastic models in the analysis and forecasting of solar radiation data by Edmore Ranganai and Mphiliseni B Nzuza was published on pages 125-137.
The corrections for the normality test in Figure 8 and the corresponding text was incorrect, and a reference was omitted. The replacement text, table and figure are as follows:
5.2 SARIMA and HCSARIMA modelling
Both SARIMA and HCSARIMA models with significant (p-values< 0.05) parameters were fitted on all four data series (see Appendices A and B, respectively) via maximum likelihood (ML) estimation. To check the adequacy of these models, tables of residual analysis based the Box-Ljung statistics (p-values> 0.05) for no serial correlation and; histograms (bell-shaped) of residuals, Q-Q plots (approximately straight line), and the Kolmogorov-Smirnov, Cramer-von Mises and the Anderson-Darling normality tests (p-values > 0.05) for normality of residuals. Satisfying all criteria in parenthesis constitute adequacy. For brevity, only results for model A are given in Appendix C.