Time-series Forecast of Nigeria’s Electricity Statistics From 1991-2028 Using Auto-regressive Integrated Moving Average (Arima) Model
Chapter One
Objectives of the Project
- Carry out a statistical forecast of the energy data of Nigeria between the year 1991 and 2028 (38 years) using an efficient model.
- Predict based on the forecast obtained, the projected electricity production, consumption and transmission losses and demand estimate of the nation in the short-
CHAPTER TWO
LITERATURE REVIEW
Electricity Demand In Nigeria
Demand is an economic term that means the quantity of a product or service that a consumer shows willingness and ability to pay a specific price for at a specific time. Demand can change with time and price. Due to increase in population, increased industrialization and the efforts of nations to reduce the combustion of fossil fuels the world faces a steady rise in demand for alternative energy sources (Madueme, 1979) and specifically electricity as a form of energy to replace the fossil fuels and that includes African nations like Nigeria.
Electricity is a si-ne-qua-non, a necessary requirement for the operation of production processes from design to manufacturing of goods, distribution and services. It plays a very important role in the socio-economic and technological development of every nation (Adebanjo, 2012).
There are indications that the nation is still far away from becoming self-sufficient in electricity, going by revelations by the Minister for Power in Nigeria, Prof Chinedu Nebo that at a time Nigeria has shifted the target date for generating 10,000 mega watts from December 2013 to the first quarter of 2014, about 200,000MW is needed to meet current energy demands.
Electrical energy is demanded in Nigeria mainly by households who serve as the final consumers because of the low level of industrialization in the country. However, the manufacturing industry, educational institutions, health care parastatals and the private and public sector firms also consume a good percentage of the electricity generated (Abatan, 1979).
Globally, electricity finds increased usage as the drive to reduce dependence on fossil fuels increases, automobile manufacturers are producing electric cars, the steel making and metallurgical industry is tending towards adoption of more electric arc furnace systems instead of the traditional blast furnace system which uses mainly carbon-based fuels and coking materials.
Although, Nebo failed to give a timeline on when the power need would be met, he told a group of British businessmen led by the Lord Mayor of the City of London, Roger that the target presents huge investment opportunity for local and foreign firms.
Nigeria currently generates around 4,500mw of electricity, and, according to him, investment particularly in the nation‘s power generation segment is largely dominated by firms from Asia and the United States of America.
He explained that the Nigerian government is determined to diversify its energy mix by increasing power generated from wind, solar and other bio-mass to reduce dependence on gas and hydro powered plants.
According to the Minister, ―we need a robust energy mix, for security purposes, as a nation it is not expedient to have a source or two we need a good mix of power, we now have wind turbines, solar power too are being commissioned, but we are still far considering our potentials. Only recently a South Korean firm has agreed to produce for us 1,000 solar panels every year for the next 10 years‖.
According to (Ibidapo-Obe and Ajibola, 2011), some rural parts of Nigeria are known to be good for wind and this robust power mix of wind and solar energy will be good for our off- grid rural folks, who will intend to focus under the Rural Electrification Scheme.
Sources and methods of compiling electricity supply and demand statistics in Nigeria
The Power Holding Company of Nigeria (PHCN) is virtually the only source of statistical information on electricity supply and demand in Nigeria.
Each of the Power Stations completes Form 125 titled ―Generation Returns‖. Information supplied includes:
- Name of Power
- Number of generating
- Energy
- Turbine
- Quantity of electricity generated in gigawatt
- Installed
Each of the Zonal Offices also completes sales returns on which, among others, information is supplied as summarised for all consumers as follows:
- Consumer category (domestic, commercial, industrial),
- Unit consumed and
- Amount paid.
CHAPTER THREE
MATERIALS AND METHODS
Research Statement and Research Questions
The Nigerian energy sector has not been able to provide adequately for the demand of electrical energy by the Nigerian public. In actual fact, there is a problem of lack of ability to coordinate data collection to generate a forecast of the actual energy statistics of the country. This paper wishes to achieve a thorough analysis of statistical data that shows the energy demand of the Nigerian economy and through a process of statistical forecasting, achieve the recognition of a known pattern of energy demand with time and hence generate a time-series forecast for the energy demand for subsequent years.
As long as energy supply cannot meet the energy demand of the people, there will be little hopes of achieving enviable national development and even improving the standard of living of the Nigerian people. The onus lies on the management personnel of the energy regulatory body, NERC to be able to obtain specific information that denotes the energy demand of the people and be able to collate such information for energy demand planning by policy makers and generation companies and other relevant stakeholders.
Materials Used
The materials used in the study included
- Microsoft Word Processing Application (MicrosoftWord),
- Statistical Models; ARIMA – Auto Regressive Integrated Moving
- IBM SPSS Data Analysispackage
Data Collection and Gathering Techniques
Statistical and numerical data was collected mainly by extensive survey of previous literature available about the subject. The survey method predominantly involves reading of such literature as books, journals, articles and bulletins of organizations with interests relevant to the topic. Some of the relevant organizations whose bulletins and reports were used in this research as primary sources include:
- World Bank‘s World Development Indicator Bulletin,
- Energy Commission ofNigeria
- Central Bank of
Time series data for the period 1980-1999 on electricity production, electricity consumption, electricity consumption per capita, transmission and distribution losses in Nigeria were used for the forecast. These data were obtained from the Central Bank of Nigeria (CBN) Statistical Bulletins for the year 1999 and 2011 and World Bank Bulletin, 2012.
CHAPTER FOUR
DATA PRESENTATION
Data Presentation
The following datum needed for complete analysis and forecasting were obtained from relevant agencies and sources stated below:
Statistical Bulletin of the Energy Commission of Nigeria (2006).
National Development Indicators Data of Nigerian Bureau of Statistics, 2012. 1999 Statistical Bulletin of the Central Bank of Nigeria.
2012 Statistical Bulletin of the Central Bank of Nigeria. 2012 World Development Indicator Bulletin of World Bank.
CHAPTER FIVE
CONCLUSION AND RECOMMENDATIONS
Conclusion
In this research work, I have been able to work with the IBM SPSS application and several packages under it like the ChartBuilder and the Time Series Modeller to create a statistical Time-series forecast of the electricity production, consumption, demand estimate and transmission and distribution losses in Nigeria having obtained the absolute figures of these parameters between the year 1991 and 2012 from the relevant agencies within and outside the country. I have subjected my forecasts to the approved tests of accuracy signified by the MAPE, RMSE and other error estimates. Forecasts are deemed to be accurate and authoritative enough if the MAPE value is below 10 and a forecast whose MAPE is between 1 and 5 is considered authoritative and accurate. In this forecast, I was able to achieve a MAPE of 1.207 for the forecasts below the 10th Percentile of all forecasts and a MAPE of 1.714 for the forecasts between the 10th and 25th Percentile of all forecasts, which is the forecast between 2012 and 2017. This value increased to 25.037 as the model generated more forecasts. Hence, the forecasts between 2012 and 2017 are very accurate and authoritative. Forecast accuracy generally diminishes as the number of forecasts generated increases and the farther away the forecasts get from the observations and these accounts for the increase in MAPE value. Averagely, the MAPE for the complete forecast stands at 8.750. Also, the RMSE value at forecasts below the 10th Percentile is 1.043 and for values at the 25th and 50th Percentile are 1.192 and 2.026 respectively. This value increased to 12.405 as the model generated more forecasts. The mean RMSE is placed at 4.221. Based on this error estimates obtained, the forecast can be adjudged to be fairly accurate and the model used has proved to be acutely capable of carrying out these forecasts if we consider the prospects of its usability in helping our electricity regulatory commission and other relevant statistical agencies in determining how to manage data that gives a good picture of the electricity demand, production and consumption patterns that help in determination of management processes of the supply chain of electricity.
Recommendation
I will recommend that the results of these forecasts be considered by the NERC as a possible indicator of the direction in which the Nation‘s energy statistics are tending to. Also, I will recommend that the model used to carry out this forecast be considered for use in the nation‘s regulatory organisations because if it can create a forecast with the level of accuracy it has done, then a trained statistician in the Nigerian Bureau of Statistics, for instance, may have an idea about how to customize it‘s working formulas to create a customised model that can work more accurately for the intricacies that form our national statistics. This has been done by the United Nations Environment Programme and the model used is now called the Egain Forecasting Model.
References
- Abatan, A.O. (1979). Optimizing Energy Utilization and Conservation. A Presentation submitted to the Federal Government of Nigeria on Fundamental Energy Policy Considerations Page 711 – 720.
- Ajayi, O.O. (2006). Mainstreaming Statistics with Policy Processes and National Development Programmes. A Paper presented to the National Strategies for the Development of Statistics, November 21 2006.
- Akinlo, A. (2008). A Study of the relationship between energy consumption and economic growth for eleven countries in Sub-Saharan Africa using the Auto-Regressive Distributed Lag Bounds test. Retrieved March 16, 2014 from www.energysustainsoc.com
- Anaekwe, C. (2010). European Scientific Journal February 2013 9(4). ISSN: 1857 – 7881
- (Print), e – ISSN 1857- 7431 28).
- Bozarth, C. (1998). Forecasting Principles: What you need to know About Forecasting. Retrieved February 28, 2014 from www.forecastingprinciples.com
- Bozarth, C. (2011). Measuring Forecast Accuracy: Approaches to Forecasting. Retrieved March 13, 2014 from www.forecastingprinciples.com
- Central Bank of Nigeria. (1999). Statistical Bulletin of the Central Bank of Nigeria. Retrieved March 15, 2014 from www.cbn.gov.ng
- Central Bank of Nigeria. (2012). Statistical Bulletin of the Central Bank of Nigeria. Retrieved March 15, 2014 from www.cbn.gov.ng
- Cheng, C. and Lai. (1997). Investigation of Relationship between energy and GNP and energy and employment to Taiwanese data for 1955 to 1993 periods. Retrieved February 2, 2014 from www.books.google.com
- Clements, R. (2010). The Absolute Best Way to Measure Forecast Accuracy. Axsium Retail Forecasting Series, October 10, 2010 Episode.
