الفهرس 
Chapter OneOnly 14 pages are availabe for public view
 |  | from | 218 |  |  |
| | | from | 218 | | |
Abstract
Introduction
Egypt is a hydrocarbon production pioneer, as it is one of the oldest energy producers and refiner in the Middle East & North Africa region. The country is considered the largest oil producer in Africa out of the Organization of the Petroleum Exporting Countries (OPEC) and the third largest natural gas producer on the continent following Algeria and Nigeria, according to the Energy Information Administration. The country is also an active member of the Organization of Arab Petroleum Exporting Countries (OAPEC) since 1973.
Egypt has significant energy resources, both in fossil fuels (oil and natural gas) and in renewable energy. Depending on fossil fuels as it is the main source of energy, the country has witnessed an increasing level of oil and natural gas consumption. Those two main energy sources (oil & natural gas) represent most of the energy consumption in Egypt, accounting for more than 95% of total energy consumption. Egypt started turning to natural gas to replace oil in the domestic market, mainly for fuelling heavy industries and electric power plants to save more crude oil for exports. So this study aims to statistically analyze the causal relationship between petroleum consumption and economic growth in Egypt within a multivariate time series framework.
Importance of Study:
The oil and gas industry in Egypt is one of the main pillars of the national economy; it is the main source of energy and a significant source of revenues. It plays a key role in satisfying the energy required for social, industrial, economic development plans, it encourages domestic production and reduces imports, along with rationalizing foreign currency utilization to maximize the surplus and achieve optimum monetization of natural resources. Thus this study aims to statistically analyze the relationship between petroleum consumption and economic growth in Egypt over the period 2001 Q3–2019Q4. Petroleum consumption will be decomposed into Gasoline, Fuel Oil, Gas Oil, Kerosene, LPG (Liquefied Petroleum Gas) and Natural Gas, in order to measure the impact of each individual components of energy on domestic production and hence on economic growth. The issue is debatable in Egypt as to determine which source of energy should be utilized to sustain economic growth.
Objectives of Study:
The main objectives of this study are:
To build statistical model to predict the causal relationship between petroleum consumption and economic growth in Egypt within a multivariate time series framework by including measures for capital, labor, Export and financial development in the aggregate production function.
The analysis is disaggregated by different fuel types including Gasoline, Fuel Oil, Gas Oil, Kerosene, LPG and Natural Gas, to account for any potential aggregation bias and to measure the impact of each individual components of energy on domestic production.
To use the Autoregressive Distributed Lag (ARDL) Model and Error Correction Model (ECM) to build the statistical models and to evaluate the dynamic interactions and strength of causal relations among variables in the system.
To use the support vector machine for performing regression analyses on time series data and to statistically analyze the causal relationship between petroleum consumption and economic growth in Egypt.
To construct feed-forward back propagation neural network and use it to investigate the causal relationship between petroleum consumption and economic growth in Egypt.
Data Sources:
All data have been obtained from the
Egyptian General Petroleum Corporation (EGPC)
Egyptian Natural Gas Holding Company (EGAS)
The Central Bank of Egypt (CBE)
Ministry of Planning, Monitoring and Administrative Reform
Central Agency for Public Mobilization and Statistics (CAPMS)
Variables of the Model:
Dependent Variable
Economic growth: The real gross domestic product series in the national currency is used as indicator for economic growth.
Independent Variable
Gross fixed Capital Formation
Employed labor
Exports
Financial Development (M2)
Petroleum Consumption will be disaggregated into :
Gasoline Consumption
Fuel Oil Consumption
Gas Oil Consumption
Kerosene Consumption
LPG (liquefied Petroleum Gas)Consumption
Natural Gas Consumption
Organization of study:
This study is presented in 6 main chapters summarized as follows:
Chapter One is an introductory chapter that provides an overview of the study. It includes an introduction, statement of the problem, importance of study, objectives of study, and limitations of study. It also includes a brief literature review; the sources of data, variables of the model are also presented in this chapter.
Chapter Two This chapter will provide a description of the main features of Egypt’s oil and gas industry with special focus on production, consumption and Exports of Oil & Gas. As well as analytical review of petroleum industry structure and refining & refined oil products.
Chapter Three deeply will discuss the different multivariate statistical techniques including: Autoregressive Distributed Lag (ARDL) approach to Co-integration and Error Correction Model (ECM), Vector Autoregressive Model (VAR), and Vector Error Correction Model (VECM).
Chapter Four This chapter will focus on the machine learning technique used in the study which are Support Vector machines and Feed Forward Neural Network.
Chapter Five This chapter will apply the various statistical techniques and the results will be analyzed and discussed.
Chapter Six This chapter will present summary of findings, conclusions, recommendations and future work.
Results:
Results of ARDL approach to cointegration
The study used the ARDL bounds testing and Gregory and Hansen structural break cointegration approaches for long run while stationarity properties of the variables have been tested applying ADF, PP and KPSS unit root tests. Also Perron–Vogelsang structural break unit root test is conducted to take into account the structural change in the variables.
Natural Gas Model
The estimated coefficients of the long-run relationship show that natural gas consumption in Egypt has a significant positive effect on economic growth. A 1 unit increase in ln natural gas consumption leads to approximately 0.327 unit increase in ln economic growth, keeping other things constant.
The equilibrium correction coefficient 〖ECM〗_(t-1), is estimated and equal to -0.685 which is negative and statistically significant at 1 percent level of significance. This means that 68.5 % of the disequilibrium in the economic growth function for the current period will be corrected in the following quarter.
.Gasoline Model
The estimated coefficients of the long-run relationship show that gasoline consumption in Egypt has a significant positive effect on economic growth. A 1 unit increase in ln gasoline consumption leads to approximately 0.232 unit increase in ln economic growth, keeping other variables constant.
The equilibrium correction coefficient 〖ECM〗_(t-1), is estimated and equal to -0.967 which is negative & statistically significant at 1 percent level of This means that 96.7 % of the disequilibrium in the economic growth function for the current period will be corrected in the following quarter.
Gas Oil Model
The estimated coefficients of the long-run relationship show that gas oil consumption in Egypt has a significant positive effect on economic growth. A 1 unit increase in ln gas oil consumption leads to approximately 0.331 unit increase in ln economic growth, keeping other variables constant.
The equilibrium correction coefficient 〖ECM〗_(t-1), is estimated and equal to -0.792 which is negative & statistically significant at 1 percent level of significance. This means that 79.2 % of the disequilibrium in the economic growth function for the current period will be corrected in the following quarter.
Fuel Oil Model
The estimated coefficients of the long-run relationship show that fuel oil consumption in Egypt has a significant positive effect on economic growth. A 1 unit increase in ln fuel oil consumption leads to approximately 0.044 unit increase in ln economic growth, keeping other variables constant.
The equilibrium correction coefficient 〖ECM〗_(t-1), is estimated and equal to -0.791 which is statistically significant at 1 percent level of significance. This means that 79.1 % of the disequilibrium in the economic growth function for the current period will be corrected in the following quarter.
Kerosene Model
The estimated coefficients of the long-run relationship show that Kerosene consumption in Egypt has a negative effect on economic growth but it is insignificant.
The equilibrium correction coefficient 〖ECM〗_(t-1), is estimated and equal to-0.5019 which is statistically significant at 1 percent level of significance and has the correct sign. This means that 50.19 % of the disequilibrium in the economic growth function for the current period will be corrected in the following quarter.
LPG Model
The estimated coefficients of the long-run relationship show that LPG consumption in Egypt has a positive effect on economic growth but it is insignificant.
The equilibrium correction coefficient 〖ECM〗_(t-1), is estimated and equal to -0.507 which is statistically significant at 1 percent level of significance. This means that 50.7 % of the disequilibrium in the economic growth function for the current period will be corrected in the following quarter.
Results of Support Vector Regression
The results indicated that SVM provides a promising alternative for time series forecasting. For natural gas model MSE in training set =0.005 and in validation set MSE=0.001. For gasoline model MSE in training set =0.003 and for validation set MSE=0.008. For gas oil model MSE in training set =0.004 and in validation set MSE=0.001. For fuel oil model MSE in training set =0.003 and in validation set MSE=0.002. For Kerosene model MSE in training set =0.003and in validation set MSE=0.01. Finally, for LPG model MSE in training set =0.006 and for validation set MSE=0.001, which indicate very high accuracy of the predicted models.
Results of Feed Forward Neural network
The results of this study indicated that FFNN showed significant results in dealing with the data.
For natural gas model MSE in training set =0.000738 and in testing set MSE=0.005636. For gasoline model MSE in training set =0.000437 and for testing set MSE=0.008352. For gas oil model MSE in training set =0.00071 and in testing set MSE=0.005675. For fuel oil model MSE in training set =0.000807 and in testing set MSE=0.008753. For Kerosene model MSE in training set =0.00091and in testing set MSE=0.004312. Finally, for LPG model MSE in training set =0.000649 and for testing set MSE=0.00664, which indicate very high accuracy of the predicted models.