Theories and empirical researches about the relationships between energy use, population growth, and GDP per capitaCO2 emissionsChapter 1: Literature review about the relationship between CO2 emissions and GDP per capita, population growth, energy use, forest area, industry and renewable energy consumption.Chapter 2: DataChapter 3: Statistics description of variables Chapter 4: Quantitative analysisChapter 5: Conclusion and policy implication
LITERATURE REVIEW
Theories
1.1 CO 2 emissions (metric tons per capita) a Definition and roles of CO 2 (Carbon dioxide)
Carbon dioxide (CO2) is a colorless gas that is approximately 60% denser than dry air It is composed of one carbon atom covalently double bonded to two oxygen atoms and is found naturally in trace amounts within the Earth's atmosphere.
CO2 plays a crucial role on Earth as plants utilize it during photosynthesis to create carbohydrates, which are essential for food This process is vital for the survival of early life, as humans and animals depend on plants for sustenance.
CO2 emissions can have detrimental effects, contributing to climate change as carbon dioxide accumulates in the atmosphere Additionally, indoor CO2 levels can rise rapidly beyond recommended limits, leading to adverse health effects.
CO2 emissions primarily stem from the combustion of fossil fuels and the production of cement This includes carbon dioxide released from solid, liquid, and gaseous fuels, as well as gas flaring that occurs during fuel consumption.
This research identifies six key factors influencing CO2 emissions: energy use, population growth, GDP per capita, forest area, industrial activities, and renewable energy consumption Energy use, particularly per capita consumption, plays a critical role in determining emission levels.
Energy use is the amount of energy or power used (kg of oil equivalent per capita). b Population growth
Population growth refers to the rise in the number of individuals residing in a specific area, such as a nation or state It can be calculated using the formula: (birth rate + immigration) - (death rate + emigration) This information is crucial for businesses and government entities when making investment decisions in various populations or regions Additionally, GDP per capita is an important metric related to this growth.
GDP per capita measures a country's economic performance by dividing its gross domestic product (GDP) by its total population This calculation provides insight into the average economic output per person, reflecting the standard of living and economic health of a nation.
3 population That makes it a good measurement of the standard of living of a country It tells you how prosperous a nation feels for each of its people. d Forest area (sq Kilometre)
Forests serve as crucial habitats for diverse plant species, playing a vital role in carbon dioxide absorption through photosynthesis and acting as significant carbon storage systems As the area of forests fluctuates, so does the capacity of these carbon pools, directly impacting the levels of CO2 released into the atmosphere.
Industry has the most impact on the environment in all remaining industries, especially for developing countries, which are in the process of Industrialization - Modernization of the country.
As industrial value increases, so does the need for higher production levels; however, many countries lack stringent regulations for exhaust gas treatment, resulting in rising emissions from factories, with CO2 being the primary pollutant Consequently, the industrial sector significantly impacts atmospheric CO2 levels, highlighting the urgent need for increased renewable energy consumption to mitigate these effects.
The rise of renewable energy sources offers a viable solution to decrease dependence on traditional energy, while also enhancing economic performance Increased utilization of renewable energy significantly improves environmental quality by lowering CO2 emissions, contrasting with the detrimental effects of conventional energy sources that contribute to higher emissions.
Theories and emperical researches about the relationships between energy use, population growth, GDP per
2.1 Energy use and CO2 emissions
Sustainable development (SD) focuses on balancing economic and social progress with environmental protection, often referred to as the 'Three Pillars' model It acknowledges that Earth's physical resource limitations and the capacity to absorb pollutants will shape our future, influenced by natural laws and human innovation SD serves as a pathway to achieving sustainability, particularly in the context of energy consumption and pollution, drawing from various fields such as architecture, economics, ecology, and social sciences.
Thao and Chon highlight that while energy consumption positively impacts the economy, it adversely affects the environment, particularly through fossil energy use, which is a major contributor to global warming and climate change The detrimental environmental effects stem not only from energy consumption but also from extraction methods In contrast, renewable energy consumption is inversely related to CO2 emissions, indicating that increasing its use can reduce emissions Ito's research further reveals that fossil energy consumption negatively affects economic growth in developing countries, whereas renewable energy promotes growth The harmful emissions from fossil fuels contrast with the more environmentally friendly residues of renewable energy Additionally, Shafei and Ruhul's analysis of the Kuznets Curve Hypothesis indicates that non-renewable energy consumption correlates positively with CO2 emissions, reinforcing the idea that boosting renewable energy use can significantly lower emissions.
2.2 Population growth and CO 2 emissions
Ehrlich (1968) and Holder and Ehrlich (1974) introduced the IPAT equation, which illustrates the relationship between demographic changes and environmental stress by linking population size, income, and environmental impact per unit of economic activity This equation serves as a valuable framework for assessing anthropogenic environmental changes, particularly in relation to CO2 emissions influenced by population, income, and technology.
The relationship between population pressure and environmental quality has been a longstanding debate, beginning with Malthus (1798 [1970]), who argued that rising population growth exerts pressure on limited land resources, potentially outpacing food production and leading to reductions through poverty, disease, famine, and war In contrast, Boserup (1981) posited that high population density can drive technological advancements in agriculture, suggesting that population growth may stimulate innovation rather than hinder sustainability.
Technological advancements have significantly enhanced food production and distribution, enabling the natural environment to support a larger population while maintaining a high standard of living.
Population growth significantly impacts environmental quality, as each individual requires energy for essential needs such as food, water, clothing, and shelter According to Birdsall (1992), this increase in population can lead to higher energy demands for power, industry, and transportation, ultimately resulting in greater fossil fuel emissions Additionally, population growth can exacerbate greenhouse gas emissions through deforestation, as it contributes to tree loss, alterations in land use, and the burning of fuel wood.
Two key questions need to be addressed: Firstly, does population pressure significantly influence carbon dioxide emissions while holding affluence and technology constant? Secondly, has the impact of demographic pressure been more pronounced in developing countries compared to developed nations?
2.3 GDP per capita and CO 2 emissions
The Environmental Kuznets Curve hypothesis suggests that wealthier nations have historically contributed to environmental degradation at a quicker pace than developing countries However, recent analyses have reevaluated the link between economic growth and ecosystem protection, challenging the traditional views on this relationship.
The theory posits that as the economy declines to a specific average income level, economic growth will eventually lead to reinvestment in the environment, facilitating the restoration of the ecosystem.
Critics argue that economic growth doesn’t always lead to a better environment and sometimes the opposite may actually be true.
In her 2012 paper, Bratt examines three distinct theories regarding the relationship between environmental degradation and GDP: the Environmental Kuznets Curve (ECC), the Brundtland Curve, and the Daly Curve While all three theories acknowledge that GDP impacts environmental degradation, they present differing perspectives The ECC suggests an inverted U-shaped relationship, where pollution increases with GDP until a certain point, after which it declines Conversely, the Brundtland Curve indicates a U-shaped correlation, showing that both the poorest and richest countries experience higher pollution levels The Daly Curve, however, posits a continuous rise in emissions alongside GDP growth without a turning point Bratt emphasizes that these curves address various forms of environmental degradation, with the ECC applicable for emissions calculations, the Brundtland Curve for output measurement, and the Daly Curve for consumption assessment Ultimately, Bratt concludes that, despite the validity of each curve, a positive and continuous relationship between environmental degradation and GDP is the most likely scenario.
This thesis explores the relationship between CO2 emissions and three key factors: per capita GDP, energy usage, and population growth While numerous studies have investigated CO2 in relation to individual factors, there is a scarcity of research that simultaneously analyzes all three influences on CO2 emissions Utilizing current data and linear regression analysis, this study aims to fill that gap and provide a comprehensive understanding of how these variables interact with CO2 emissions.
2.4 Forest area and CO 2 emissions
Deforestation, particularly of tropical forests, significantly contributes to rising CO2 levels, accounting for approximately 23% of human-emitted CO2 emissions Trees and plants play a crucial role in the carbon cycle by absorbing CO2 and converting it into essential nutrients Rather than just reducing atmospheric CO2, forests serve as vital carbon reservoirs; their destruction releases stored carbon back into the atmosphere, exacerbating climate change Moreover, studies, such as those by Adedire (2002), indicate that deforestation leads to reduced rainfall, increased surface temperatures, and altered local hydrology.
The significant rise in atmospheric carbon dioxide (CO2) levels is largely due to the drastic reduction or loss of trees essential for sequestration This increase in CO2 contributes to climate change and global warming, leading to severe consequences such as coastal flooding, alterations in food chains, and disruptions in agricultural production.
In the process of Industrialization - Modernization along with the socio
As economic development progresses, the establishment of factories and industrial zones has surged, leading to significant environmental challenges Many of these industrial areas still rely on outdated production technologies and lack adequate exhaust gas treatment systems, resulting in the release of substantial amounts of industrial waste gas This situation has severely impacted air quality, particularly contributing to elevated CO2 levels in the atmosphere.
Human-produced CO2 emissions, while smaller than natural gas emissions, have significantly disrupted the pre-Industrial Revolution carbon cycle Since the onset of the Industrial Revolution, artificial CO2 emissions have surged, primarily due to fossil fuel combustion and deforestation Approximately 87% of these emissions stem from burning fossil fuels like coal, natural gas, and oil, while the remaining 13% arises from deforestation, land use changes (9%), and various industrial processes (4%).
2.6 Renewable energy consumption the empirical studies by Bửlỹk and Mert (2014), Dogan and Seker (2016a, 2016b) Irandoust
(2016), Jebli et al (2016), Liu et al (2017) and Sebri and Ben-Salha (2014) incorporated RE consumption as an additional variable to explore the linkages between non-
DATA
Methodology in collecting data
The study utilizes secondary mixed data sourced from the World Bank, focusing on key factors influencing CO2 emissions per capita, measured in metric tons These factors include forest area, GDP per capita, the percentage of industry (including construction) in GDP, total population, the share of renewable energy consumption in total energy use, and per capita energy consumption.
Methodology in processing data
Using Gretl in order to process data cursorily then calculate the correlation matrix among variables.
Data overview
- The dataset was collected from the official website of World Bank, including 199 observations of 199 countries in 2015
- Data source: https://data.worldbank.org/
Forest area https://data.worldbank.org/indicator/AG.LND.FRST.K2
GDP per capita https://data.worldbank.org/indicator/NY.GDP.PCAP.CD
Industry proportion https://data.worldbank.org/indicator/NV.IND.MANF.ZS
Population https://data.worldbank.org/indicator/SP.POP.TOTL
Renewable energy consumption https://data.worldbank.org/indicator/EG.FEC.RNEW.ZS
Energy consumption per capita https://data.worldbank.org/indicator/EG.USE.PCAP.KG.OE The structure of Economic data: cross-sectional data
Data description
Run the command “des lnco2 fa gpdppercpt ind lnpop rec lnecpercp” to interpret the dataset
des lnco2 fa gdppercpt ind lnpop rec lnecpercp storage display value variable name type format label variable label
The article discusses various metrics related to environmental and economic factors, including the natural algorithm of CO2 levels, forest area, GDP per capita, industrial contributions, and population dynamics It highlights the importance of energy consumption per capita as a critical indicator of sustainability Understanding these elements is essential for analyzing the relationship between economic growth and environmental impact.
By using DES, we know clearly about the variables According to the results, we know:
lnco2: Natural logarithm of CO2 emissions (metric tons per capita)
fa: Forest area (sq km)
gpdppercpt: GDP per capita (current US$)
ind: Industry (including construction), value added (% of GDP)
lnpop: Natural logarithm of population
rec: Renewable energy consumption (% of total final energy consumption)
lnecpercp: Natural logarithm of energy consumption per capita (kWh)
STATISTICS DESCRIPTION OF VARIABLES
Using Gretl to bring out regression models by using Ordinary Least Squares method (OLS) to estimate the parameter of multi-variables regression models As a result, we can:
- Depend on variance inflation factor (VIF) to identify multicollinearity
- Use Jacque – Bera test to test Normality of residual
- Use White test and Breush – Pagan test to test Heteroscedasticity
Use Ramsey test to test for omitted independent variable
Base on related public researches and economic theories, the model used in this report is constructed to examine the impacts of relevant factors including forest area, GDP per capita,
% industry (including construction) of GDP, population in total, % renewable energy consumption of total final energy consumption, and energy consumption per capita co2 = f(fa, gpdpercpt, ind, pop, rec, ecpercpt)
co2: CO2 emissions (metric tons per capita)
fa: Forest area (sq km)
gdppercpt: GDP per capita (current US$)
ind: Industry (including construction), value added (% of GDP)
rec: Renewable energy consumption (% of total final energy consumption)
ecpercpt: Energy consumption per capita (kWh)
To display the relationship between the dependent variable, which is co2, and independent variables, fa, gdppercpt, ind, pop, rec, ecpercp, the function has the following form:
(PRF): lnco 2= β 0 + β 1 fa + β 2 gdppercpt + β 3 ind + β 4 lnpop+ β 5 rec + β 6 lnecpercp+u
(SRF): lnco 2=^ β 0 + ^ β 1 fa + ^ β 2 gdppercpt + ^ β 3 ind + ^ β 4 plnpop + ^ β 5 rec + ^ β 6 lnecpercpt + ^ u
The regression model analyzes the dependent variable lnco2, influenced by several independent variables: fa, gpdppercpt, ind, lnpop, rec, and lnecpercp The intercept term is represented by β0, while β1 to β6 denote the regression coefficients for each independent variable Additionally, u represents the disturbance within the model, capturing the unexplained variability This structured approach aids in understanding the relationships between carbon dioxide emissions and various economic and demographic factors.
^ β 0 : the estimator of the intercept
^ β 6: the estimator of β 6 u ^: the residual (the estimator of u )
The Unit lnco2 represents the natural logarithm of CO2 emissions measured in metric tons per capita Additionally, fa indicates the forest area in square kilometers, while gpdppercpt refers to GDP per capita in current US dollars Furthermore, ind encompasses the value added by the industry, including construction, as a percentage of GDP.
+ % lnpop Natural logarithm of population + people rec Renewable energy consumption
(% of total final energy consumption)
- % lnecpercp Natural logarithm of energy consumption per capita
SUM function lets us know about observations, mean, standard deviation, max and min value of the variables sum co2 fa gdppercpt ind pop rec ecpercp
Variable | Obs Mean Std Dev Min Max
-+ - co2 | 199 4.712814 5.864656 04 41.64 fa | 199 232074.4 894664.2 0 8149305 gdppercpt | 199 15514.21 22273.06 305.55 167290.9 ind | 199 25.07467 11.73998 4.15 63.84 pop | 199 3.84e+07 1.41e+08 12475 1.37e+09
By using SUM function, with 199 observations, we have:
co2: the mean CO2 emissions among 199 countries in 2015 is 4.712814, standard deviation is 5.864656, the minimum figure is 0.04 and the maximum one is 41.64.
fa: the mean forest area is 232074.4, standard deviation is 894664.2, the minimum figure is 0 and the maximum one is 8149305
gdppercpt: the mean GDP per capita is 15514.21, standard deviation is 22273.06, the minimum figure is 305.55 and the maximum one is 167290.9.
ind: the mean percentage of industry (including construction) of GDP is 25.07467, the standard deviation is 11.73998, the minimum figure is 4.15 and the maximum one is 63.84.
pop: the mean population is 3.84.10 7 , the standard deviation is 1.41.10 8 , the minimum figure is 12475 and the maximum one is 1.37.10 9
rec: the mean percentage of renewable energy consumption in total final energy consumption is 30.02769 , the standard deviation is 28.14265, the minimum figure is 0 and the maximum one is 95.82.
ecpercpt: the mean energy consumption per capita is 26670.12, the standard deviation is 35354.3, the minimum figure is 99 and the maximum one is 221634.
Run the command: corr lnco2 fa gdppercpt ind lnpop rec lnecpercp
| lnco2 fa gdpper~t ind lnpop rec lnecpe~t
-+ - lnco2 | 1.0000 fa | 0.1396 1.0000 gdppercpt | 0.5031 0.0816 1.0000 ind | 0.2277 0.0335 -0.0403 1.0000 lnpop | -0.1448 0.3369 -0.2327 0.2965 1.0000 rec | -0.7668 -0.0388 -0.2774 -0.1028 0.2210 1.0000 lnecpercpt | 0.9380 0.1354 0.5392 0.2190 -0.1855 -0.6839 1.0000
Correlation between dependent variable and independent variables:
The correlation coefficient between lnco2 and fa is 0.1396, which is quite postitive Therefore, fa has a positive effect on co2 but only a huge change in fa can shift the co2.
The correlation coefficient between lnco2 and gdppercpt is 0.5031, which is a strong correlation Therefore, gdppercpt has a positive effect on co2
The correlation coefficient between lnco2 and ind is 0.2277, which is quite postitive Therefore, ind has a positive effect on co2 but only a huge change in ind can shift the co2.
The correlation coefficient between lnco2 and lnpop is -0.1448, which is quite negative Therefore, pop has a negative effect on co2 but only a huge change in pop can shift the co2.
The correlation coefficient between renewable energy consumption (rec) and CO2 emissions (lnco2) is -0.7668, indicating a strong negative relationship This suggests that changes in renewable energy consumption significantly impact CO2 emissions, with even slight variations in rec leading to substantial alterations in CO2 levels.
The correlation coefficient of 0.9380 between lnco2 and lnecpercpt indicates a strong positive relationship This suggests that as renewable energy consumption (lnecpercpt) increases, CO2 emissions (lnco2) decrease significantly, highlighting the impactful role of renewable energy in reducing carbon emissions.
In summary, all the correlations above are appropriate with theories.
The correlation coefficient between fa and gdppercpt is 0.0816, which is quite postitive Therefore, fa and gdppercpt has a weak correlation.
The correlation coefficient between fa and ind is 0.0335, which is quite postitive. Therefore, fa and ind has a weak correlation.
The correlation coefficient between fa and lnpop is 0.3369, which is quite postitive Therefore, fa and lnpop has a medium correlation.
The correlation coefficient between fa and rec is -0.0388, which is quite negative. Therefore, fa and rec has a weak correlation.
The correlation coefficient between fa and lnecpercpt is 0.0335, which is quite postitive Therefore, fa and lnecpercpt has a weak correlation.
The correlation coefficient between gdppercpt and ind is -0.0403, which is quite negative Therefore, gdppercpt and ind has a weak correlation.
The correlation coefficient between gdppercpt and lnpop is -0.2327, which is quite negative Therefore, gdppercpt and lnpop has a weak correlation.
The correlation coefficient between gdppercpt and rec is -0.2774, which is quite negative Therefore, gdppercpt and rec has a weak correlation.
The correlation coefficient between gdppercpt and lnecpercpt is 0.5392, which is a strong correlation Therefore, gdppercpt has a positive effect on lnecpercpt.
The correlation coefficient between ind and lnpop is 0.2965, which is quite postitive Therefore, ind and lnpop has a medium correlation.
The correlation coefficient between ind and rec is -0.1028, which is quite negative Therefore, ind and rec has a weak correlation.
The correlation coefficient between ind and lnecpercpt is 0.2190, which is quite postitive Therefore, ind and lnecpercpt has a weak correlation.
The correlation coefficient between lnpop and rec is 0.2210, which is quite postitive Therefore, lnpop and rec has a weak correlation.
The correlation coefficient between lnpop and lnecpercpt is -0.1855, which is quite negative Therefore, gdppercpt and lnpop has a weak correlation.
The correlation coefficient between rec and lnecpercpt is -0.6839, which is quite negative Therefore, gdppercpt and lnpop has a strong correlation.
QUANTITATIVE ANALYSIS
Run the command reg lnco2 fa gdppercpt ind lnpop rec lnecpercp
Source | SS df MS Number of obs = 199
- lnco2 | Coef Std Err t P>|t| [95% Conf Interval]
-+ - fa | 1.35e-08 3.82e-08 0.35 0.725 -6.18e-08 8.88e-08 gdppercpt | 2.88e-06 1.72e-06 1.67 0.097 -5.22e-07 6.28e-06 ind | 0032233 0029566 1.09 0.277 -.0026084 0090549 lnpop | 0320576 0166233 1.93 0.055 -.0007302 0648453 rec | -.0132449 0015499 -8.55 0.000 -.016302 -.0101879 lnecpercpt | 6933083 0330013 21.01 0.000 6282166 7584
According to the above result, we now have:
Variables Coefficient ^ β t p-value Confidence interval
Constant -5.922798 -14.54 0.000 [-6.726303; -5.119293] fa 1.35.10 -8 0.35 0.725 [-6.18.10 -8 ;8.88.10 -8 ] gdppercpt 2.88.10 -6 1.67 0.097 [-5.22 10 -7 ;6.28.10 -6 ] ind 0.0032233 1.09 0.277 [-.0026084; 0.0090549] lnpop 0.0320576 1.93 0.055 [-.0007302; 0.0648453] rec - 0.0132449 -8.55 0.000 [-.016302; -0.0101879] lnecpercpt 0.6933083 21.01 0.000 [.6282166; 0.7584]
We have the Sample Regression Model: lnco 2= β 0 + β 1 fa + β 2 gdppercpt + β 3 ind + β 4 lnpop+ β 5 rec + β 6 lnecpercp+u
According to the estimated result from STATA using the OLS method, we obtained the Sample Regression Function (SRF) as below: lnco2 = -5.922798 + 1.35 10- fa+ 2.88.10 -6 gdppercpt + 0.0032233ind + 0 0320576 lnpop - 0.0132449 rec + 0.6933083 lnecpercp + u
2 Testing problems of the model (Dianosing the model problems)
Specification involves selecting the right functional form for a model and determining which variables to include It is crucial to identify any omitted variables that could impact the dependent variable during this selection process As long as significant omitted variables are not included in the true model, Ordinary Least Squares (OLS) regression will yield unbiased estimates.
Conducting Ramsay’s RESET test for omission of influential variables ovtest
Ramsey RESET test using powers of the fitted values of lnco2
Ho: model has no omitted variables
Conclusion: There is no influential variable omitted from the mode
Multicollinearity arises when there are strong linear relationships among two or more explanatory variables, leading to less accurate estimators This phenomenon can obscure true parameter values, as minor alterations in input data may cause significant fluctuations in the model, including potential sign changes in parameter estimates.
Imprecise estimators can yield minimal insights into true parameters, causing significant fluctuations in the model with small alterations to the input data, potentially even reversing the signs of parameter estimates.
Using Variance Inflation Factor (VIF) to test for multicollinearity by running vif function in stata: vif
-+ - lnecpercpt | 2.78 0.359448 rec | 1.98 0.505896 gdppercpt | 1.53 0.652653 lnpop | 1.44 0.694914 ind | 1.25 0.798868 fa | 1.21 0.824957
According to this result, VIF of 6 independent values are smaller than 10, multicollinearity does not occur between one and other independent variables.
Conclusion: According to the test result, the model does not face multicollinearity phenomenon.
In classic linear regression, a fundamental assumption is that the variance of the error term (Ui) remains constant when the value of the independent variable (X) is fixed This is expressed mathematically as var(ui|Xi) = σ² for all i, indicating homoscedasticity in the model.
The inherent complexities of the health-economics relationship lead to issues in data collection and processing, resulting in a violation of the hypothesis and the presence of heteroskedasticity Consequently, the Ordinary Least Squares (OLS) method fails to produce the estimator with the smallest variance, rendering the testing results unreliable.
To detect heteroskedasticity, a graphical examination of residuals can be performed In STATA, this can be accomplished using the command `rvfplot, yline(0)`, which generates a graph that illustrates the relationship between the residuals and the fitted values of the dependent variable, lnco2.
Another way to check the model’s error is using the White test to explore relationship between squared residuals and all of the dependent variables
Conducting White test for detecting Heteroskedasticity imtest, white
White's test for Ho: homoskedasticity against Ha: unrestricted heteroskedasticity chi2(27) = 39.67
Cameron & Trivedi's decomposition of IM-test
Conclusion: the model has no heteroskedasticity
Another way to check the model’s error is using the Breush – Pagan test
Breusch-Pagan / Cook-Weisberg test for heteroskedasticity
Variables: fitted values of lnco2 chi2(1) = 0.15
Conclusion: the model has no heteroskedasticity
Residual does not have normal distribution.
The reason to explain why residual does not have normal distribution may be the sample is not large enough.
The phenomenon may result in the hypothesis test is inexact with the critical value method.
One way to test normality is graphical examination. predict u, residuals histogram u, width (20) normal
Another way to test normality is using Jacque – Bera test. sktest u
Skewness/Kurtosis tests for Normality
- joint - Variable | Obs Pr(Skewness) Pr(Kurtosis) adj chi2(2) Prob>chi2 -+ - u | 191 0.0000 0.0000 0.0000
According to the result, p-value (skewness) = 0.0000 < 0.05 => reject null hypothesis
Conclusion: data does not have normal distribution
3 Hypothesis postulated a Testing hypothesis about the regression parameter
As | T |< c α =¿ T ere ℎere is not enougℎereevidence ¿ reject null ypot esis ℎere ℎere
Confidence Interval Method β fa ∈( ^ β fa −c α se ( ^ β fa ) ; ^ β fa + c α se ( ^ β fa ) ) α =5 %=¿ c α =1.972 ¿> β fa ∈ ¿ ¿> β fa ∈ (−0.002074178847 ; 0.002979927942)
As value 0 ∈( − 0.002074178847; 0.002979927942) ¿> T ere ℎere is not enougℎere evidence ¿ reject null ypot esis ℎere ℎere
As p − value= 0.7264>α =5 % ¿> T ere ℎere is not enougℎere evidence ¿ reject null ypot esis ℎere ℎere
Forest area (fa) doesn’t have statistically significant effect on CO2 emissions
Gross Domestic Productions per Capita (gdppercpt)
As | T |< c α ¿> T ere ℎere is not enougℎere evidence ¿ reject null ypot esis ℎere ℎere
Confidence Interval Method β gdppercpt ∈(^ β gdppercpt − c α se ( ^ β gdppercpt ) ; ^ β gdppercpt +c α se ( ^ β gdppercpt ) ) α =5 %=¿ c α =1.972 ¿> β gdppercpt ∈ ¿ ¿> β gdppercpt ∈ (− 0.001268724514 ; 0.01554633705)
As value 0 ∈( − 0.001268724514 ; 0.01554633705 ) ¿> T ere ℎere is not enougℎere evidence ¿ reject null ypot esis ℎere ℎere
As p − value= 0.095>α =5 % ¿> T ere ℎere is not enougℎere evidence ¿ reject null ypot esis ℎere ℎere
Gross Domestic Productions per Capita has no statistically significant effect on CO2 emissions as α =5 %
As | T |< c α ¿> T ere ℎere is not enougℎere evidence ¿ reject null ypot esis ℎere ℎere
Confidence Interval Method β ind ∈ ( ^ β ind − c α se ( ^ β ind ) ; ^ β ind + c α se ( β ^ ind ) ) α =5 %=¿ c α =1.972 ¿> β ind ∈( 0.0032233 – 1.972 x 0.0029566 ; 0.0032233+ 1.972 x 0.0029566 ) ¿> β ind ∈(− 0.0026071152 ; 0.0090537152)
As value 0 ∈ (− 0.0026071152; 0.0090537152) ¿> T ere ℎere is not enougℎere evidence ¿ reject null ypot esis ℎere ℎere
As p − value= 0.2758 >α =5 % ¿> T ere ℎere is not enougℎere evidence ¿ reject null ypot esis ℎere ℎere
Industry has no statistically significant effect on CO2 emissions as α =5 %
As | T |< c α ¿> T ere ℎere is not enougℎere evidence ¿ reject null ypot esis ℎere ℎere
Confidence Interval Method β lnpop ∈( ^ β lnpop −c α se ( ^ β lnpop ) ; ^ β lnpop + c α se ( ^ β lnpop ) ) α =5 %=¿ c α =1.972 ¿> β lnpop ∈ (0.0320576 – 1.972 x 0.0166233 ; 0.0320576 +1.972 x 0.0166233 ) ¿> β lnpop ∈ (− 0.0007235476 ; 0.06487476)
As value 0 ∈ (− 0.0007235476; 0.06487476 ) ¿> T ere ℎere is not enougℎere evidence ¿ reject null ypot esis ℎere ℎere
As p − value= 0.0536 >α =5 % ¿> T ere ℎere is not enougℎere evidence ¿ reject null ypot esis ℎere ℎere
Industry has no statistically significant effect on CO2 emissions as α =5 %
As | T |> c α ¿> reject null ypot esis ℎere ℎere
Confidence Interval Method β rec ∈ ( ^ β rec − c α se ( ^ β rec ) ; ^ β rec +c α se ( ^ β rec ) ) α =5 %=¿ c α =1.972 ¿> β rec ∈[ ( −0.0132449 ) – 1.972 x 0.0015499 ; ( −0.0132449 )+1.972 x 0.0015499] ¿> β rec ∈(− 0.0163013028;− 0.0101884972)
As value 0 ∉( − 0.0163013028;− 0.0101884972) ¿> reject null ypot esis ℎere ℎere
As p − value= 0.0004 reject null ypot esis ℎere ℎere
- Renewable energy consumption has statistically significant effect on CO2 emissions The higher the renewable enery consumption is, the higher CO2 emissions are.
- In particular, with the sample we have, the estimated result shows that an 1 % increase in renewable enerygy consumption will decrease CO2 emissions by 0.013% on average, holding other factors fixed.
Energy consumtion per capita (ecpercpt)
As | T |> c α ¿> reject null ypot esis ℎere ℎere
Confidence Interval Method β ecpercpt ∈ (^ β ecpercpt −c α se ( ^ β ecpercpt ) ; ^ β ecpercpt + c α se ( ^ β ecpercpt ) ) α =5 %=¿ c α =1.972 ¿> β ecpercpt ∈ (0.6933083 – 1.972 x 0.0330013 ;0.6933083 – 1.972 x 0.0330013) ¿> β ecpercpt ∈ (0.6282297364 ;0.7583868636 )
As value 0 ∉( 0.6282297364 ; 0.7583868636 ) ¿> reject null ypot esis ℎere ℎere
As p − value= 0.0004 reject null ypot esis ℎere ℎere
- Energy consumption per capita has statistically significant effect on CO2 emissions The higher the enery consumption is, the higher CO2 emissions are.
The analysis indicates that a 1% increase in per capita energy consumption is associated with an average rise of 0.693% in CO2 emissions, assuming all other factors remain constant Additionally, the overall significance of the model has been tested to validate these findings.
(1 − R¿¿ 2)/(n − k − 1) ¿ , where: n: numbers of observations or sample size, n = 199 k: the numbers of variables, k = 6 ¿> F= R 2 / k
As F >c α =¿ reject null ypot esis ℎere ℎere
The overall model is statistically significant at a significant level of 5%
CONCLUSION AND POLICY IMPLICATION
Forest area (fa) doesn’t have statistically significant effect on CO2 emissions As α =5 % Gross Domestic Productions per Capita has no statistically significant effect on CO2 emissions as α=5%
Industry has no statistically significant effect on CO2 emissions as α =5 %
Population has no statistically significant effect on CO2 emissions as α =5 %
Renewable energy consumption has statistically significant effect on CO2 emissions The higher the renewable energy consumption is, the lower CO2 emissions are.
Our analysis reveals that a 1% increase in renewable energy consumption leads to an average decrease of 0.013% in CO2 emissions, assuming all other factors remain constant Additionally, energy consumption per capita has a significant impact on CO2 emissions; specifically, a 1% increase in per capita energy consumption results in an average rise of 0.693% in CO2 emissions, ceteris paribus.
In 2015, carbon dioxide (CO2) emissions surged to unprecedented levels, highlighting the urgent need for solutions to mitigate these harmful greenhouse gases, which are driving global warming, rapid ice melt, and rising sea levels A report from the National Oceanic and Atmospheric Administration (NOAA) published on March 10, 2016, revealed that the annual growth rate of atmospheric CO2 reached 3.05 parts per million (ppm), marking the largest increase in 56 years NOAA scientists caution that CO2 levels are rising at a pace not seen in thousands of years, primarily due to extreme weather events and the combustion of fossil fuels.
Six key factors influence CO2 emissions: forest area (fa), gross domestic product per capita (gdppercpt), industry (ind), population (lnpop), renewable energy consumption (rec), and energy consumption per capita (ecpercpt) In 2015, renewable energy consumption and energy consumption per capita were found to have a statistically significant impact on CO2 emissions, while the effects of industry, population, forest area, and gross domestic product per capita were not statistically significant.
To protect our environment and reduce CO2 emissions, it is crucial to limit the use of fossil fuels and explore alternative energy sources Fossil fuels, such as coal and oil, are significant contributors to the greenhouse effect As a result, there is a growing interest in eco-friendly energy options like wind, solar, tidal, and geothermal energy In the United States, many states enforce laws that mandate engineered vehicles to undergo periodic testing for exhaust emissions to ensure compliance with environmental standards.
To reduce carbon emissions and protect the environment, it's essential to minimize reliance on fossil fuels for electricity, as they produce significant CO2 Utilizing natural light, energy-efficient bulbs, and turning off appliances when not in use can help conserve energy Additionally, opting for public transportation, biking to school, and limiting the use of personal vehicles like motorbikes not only saves money but also decreases greenhouse gas emissions Embracing these practices contributes to a healthier planet and a more sustainable future.
27 can be said that the higher awareness about environmental pollution, the less amount of CO2 released into the environment
Our research analyzed the interdependent relationship between CO2 emissions, forest area, industrial activity, GDP per capita, population size, renewable energy consumption, and per capita energy usage The findings align with established economic theories and corroborate several previously published studies.
There are positive impacts of total energy consumption and industrial proportion on CO2 emissions If total energy consumption and industrial proportion increase, average CO2 emissions would increase followingly.
In contrast, renewable energy consumption has negative relationship as reduce the amount of CO2 to protect environment.
The report reflects the collective effort of our group and the knowledge acquired in class Although we faced challenges in data collection and understanding, we diligently worked to enhance our comprehension of the fundamental processes involved in running econometric models This enabled us to analyze the relationships between variables and address issues related to environmental quality development.
We sincerely thank Mrs Dinh Thi Thanh Binh for her invaluable guidance and dedication, which significantly contributed to the successful completion of our report We are committed to revising our research issues in response to her insightful comments and advice, aiming to enhance both the theoretical and practical aspects of our work.
1 Per capita energy consumption in 2015 by Our World in Data, 2016 Available at: https://ourworldindata.org/grapher/per-capita-energy-use? tab=table&time 15®ion=World [Retrieved in 10 th Dec 2020].
2 Naomi Klein (2014) This Changes Everything: Capitalism vs The Climate (Hardcover), (shelved 525 times as environment), 65-101.
3 Elizabeth Kolbert (2014) The Sixth Extinction: An Unnatural History
(Hardcover), (shelved 498 times as environment) 34-97
4 Thomas L.Friedman (2008) Hot, Flat, and Crowded: Why We Need a Green
Revolution – and How It Can Renew -America (Hardcover) (shelved 186 times as environment), 129-187
5 Greenhouse gases equivalencies calculator – calculations and references, Energy and the Environment, 2016 Available at: https://www.epa.gov/energy/greenhouse-gases- equivalencies-calculator-calculations-and-references [Retrieved at 11 th Dec 2020].
6 CO2 and Greenhouse gas emissions country profiles by Our World in Data, 2017. Available at: https://ourworldindata.org/co2-and-other-greenhouse-gas-emissions [Retrieved at 10 th Dec 2020].
7 Ralph B Alexander (Goodreads Author) (2010) Global Warming False Alarm:
The Bad Science Behind the United Nations' Assertion that Man-made CO2 Causes Global Warming 5 th edition, 25-85.
8 Mark Sloan (2020) Bath Bombs & Balneotherapy: The Surprising Health
Benefits of Bath Bombs and Ancient Secrets of Hot Springs, Dead Sea Minerals and CO2 Baths for Beautiful Skin, Increased Energy, and Weight Loss 2 nd edition, 143-176.
9 Rise of carbon dioxide unbated by NOAA research NEWS, April 2020 Available at: https://research.noaa.gov/article/ArtMID/587/ArticleID/2636/Rise-of-carbon-dioxide- unabated [Retrieved at 12 th Dec 2020].
9 GDP per capita (current US$) in 2015 by World Bank national accounts data, and
OECD National Accounts data files Available at: https://data.worldbank.org/indicator/NY.GDP.PCAP.CD [Retrieved 10 th Dec 2020]
10 Greenhouse gas emissions, the World Factbook, 2015 Available at: https://www.cia.gov/library/publications/resources/the-world-factbook/docs/ notesanddefs.html#279 [Retrieved at 10 th 2020]
11 Global carbon emissions from fossil fuel in 2015, June 2016 Available at: https:// www.co2.earth/global-co2-emissions [Retrieved at 10 th Dec 2020]
12 CO2 Emissions from Fuel Combustion: Overview, July 2016 Available at: https://www.iea.org/data-and-statistics/data-tables [Retrieved 10 th Dec 2020].
13 Tropical rainforests: Disappearing Opportunities Carbon Dioxide Emissions
Charts, 2015 Available at: https://rainforests.mongabay.com/09-carbon_emissions.htm
14 P.D Sharma (2005) Ecology and Environment 4 th edition, 15-43
15 Peter Zumthor, Brigitte Labs-Ehlert (2006) Atmospheres Architectural
Environments Surrounding Objects 12 th edition, issue 1, 23-65
16 Jeffrey M.Wooldridge (2012) Introductory Econometrics: A Modern Approach
(Upper-Level Economics Titles) 7 th edition, Regression Analysis with Cross-Sectional Data,
17 Trends in global CO2 emissions: 2015 report, 2016 Available at: https://www.pbl.nl/en/publications/trends-in-global-co2-emissions-2015-report [Retrieved at 10 th Dec 2020]
18 Joshua D.Angrist, Jorn – Steffen Pischke (2008) Mostly Harmless Econometrics:
An Empiricist’s Companion (Illustrated Edition), Making Regression Make Sense, 22-81.
19 The world's CO2 emissions fell in 2015 But don't celebrate just yet, August
2015 Available at: https://www.vox.com/2015/12/8/9873372/emissions-drop-2015 [Retrieved 10 th Dec 2020]
20 CO2 emissions (metric tons per capita): Carbon Dioxide Information Analysis
Center, Environmental Sciences Division, Oak Ridge National Laboratory, Tennessee, United States, 2016 Available at https://data.worldbank.org/indicator/EN.ATM.CO2E.PC
21 How the world’s largest green search engine is fighting climate change by
Heather Farmbrough, Nov 2020 Available at: https://www.forbes.com/sites/heatherfarmbrough/2020/11/10/fighting-climate-change-with- the-worlds-largest-green-search-engine/?sh87d88e44021 [Retrieved in 11 th Dec 2020]
22 Irina Perminova, Kirk Hatfield, Norbert Hertkorn (2005) Use of Humid
Substances to Remediate Polluted Environments: From Theory to Practice: Proceedings of the NATO advanced Research Workshop on Use of Humates to Remediate Polluted Environments 1 st edition, 143-178.
23 Jeffrey M.Wooldridge (2010) Econometric Analysis of Cross Section and Panel
Data (MIT press), Instrumental Variables Estimation of Single-Equation Linear Models 2 nd edition, 89-122.
24 Proportions of economic sectors in GDP in selected countries 2015 by Various Artists, Nov 2016 Available at: https://www.statista.com/statistics/264653/proportions-of- economic-sectors-in-gross-domestic-product-gdp-in-selected-countries/ [Retrieved at 10 th Dec 2020].
25 Renewable energy consumption (% of total final energy consumption) by World
Bank, Sustainable Energy for All in 2015, 2016 Available at: https://data.worldbank.org/indicator/EG.FEC.RNEW.ZS [Retrieved 10 th Dec 2020]
26 Environmental health in Health topics by World Health Organization, 2016. Available at: https://www.who.int/health-topics/environmental-health#tab=tab_1 [Retrieved at 12 th Dec 2020]
27 Global per capita energy consumption by select country 2015 by Ian Tiseo on
Energy & Environment, Statista, 2020 Available at: https://www.statista.com/statistics/268151/per-capita-energy-consumption-in-selected- countries/ [Retrieved at 10 th Rec 2020]
28 Willian H Greene, Stern School of Business, New York University (2012).
Econometric Analysis, The Linear Regression Model, 7 th edition, 41-259.
29 Peter Kennedy (February 2008) A Guide to Econometrics, The classical Linear
30 12 ways you can protect the environment by Green Mountain Energy, 2020. Available at: https://www.greenmountainenergy.com/why-renewable-energy/protect-the- environment/ [Retrieved at 12 th Dec 2020].
31 Forest area (sq km) by Food and Agriculture Organization, electronic files and web site on data world bank in 2015, 2016 Available at: https://data.worldbank.org/indicator/AG.LND.FRST.K2 [Retrieved in 10 th Dec 2020].
32 World population in 2015 by World Population Data, 2017 Available at: https:// www.prb.org/2015-world-population-data-sheet/ [Retrieved at 10 th Dec 2020]
1 The dataset of global CO2 emissions (kilo tons per capita) in 2015
CO2 emissions (tons per capita)
GDP per capita (current US$)
Industry (includin g construct ion), value added (% of GDP)
Renewable energy consumption (% total final energy consumption
Energy consumptio n per capita (kWh)
St Vincent and the Grenadines 2.02 270 6,920.88 15.41 109148 5.81 9,373