Modelling time series structure, identifying outliers and forecasting ESKOM electricity production data using singular spectrum analysis
Jacques de Klerk
What the paper says
Electricity supply by Eskom, who produces around 95% of South Africa’s electricity, has experienced a steady decline over recent years. The power utility is struggling to meet demand and South Africans are facing the brunt thereof in the form of daily load-shedding. It is of paramount importance for bulk users of electricity, e.g., mines, smelters (iron ore and aluminium), municipalities and so forth, to accurately forecast electricity supply for planning purposes. The production time series is rich with trend and seasonal features and well suited for a time series method such as Singular Spectrum Analysis (SSA). SSA can handle trends that include polynomials of any order and/or exponential trends. The method can also handle seasonality of any periodicity combined with/without trends. SSA embeds an observed time series into a so-called Hankel structured trajectory matrix and singular vector decomposition (SVD) then ensues. Singular vectors are inspected to assess possible trend and/or seasonality present in an observed times series. Once the trend and/or seasonality has been established, outlier identification and robust forecasting can ensue.
Evidence weight
Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40
| F · citation impact | 0.50 × 0.4 = 0.20 |
| M · momentum | 0.50 × 0.15 = 0.07 |
| V · venue signal | 0.50 × 0.05 = 0.03 |
| R · text relevance † | 0.50 × 0.4 = 0.20 |
† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.