Ismail, S., Zulkifli, M., Mansor, R., Mat Yusof, M. and Ismail, M. I.
Pertanika Journal of Tropical Agricultural Science, Volume 24, Issue S, October 2016
Keywords: Graphical, classical approach, time series components, line chart, moving seasonal
Published on: 09 Dec 2016
Exploratory Data Analysis (EDA) is an approach introduced by Tukey. The aim is to obtain an in depth understanding of data using graphics. This contradicts the classical approach which uses statistical summaries such as mean and standard deviation among others. Skipping EDA in investigating data can jeopardise the validity and reliability of the results. In this study, we highlight the crucial role of EDA in electricity forecasting via line chart to understand the existence of the four-time series components i.e. trend, seasonal, irregular and cyclical. Daily electricity load for financial year 2002 to 2006 are used. The EDA showed irregular patterns identified in the line chart are not irregular components but moving seasonal components which require different techniques of forecasting.
ISSN 1511-3701
e-ISSN 2231-8542