Impacts of Weather in Electricity Consumption in Sacramento Area - Essay Example

Table of Contents Table of Contents 2 Background of the Study 3 Hypotheses 3 Hypothesis 3 Hypothesis 2: 4 Hypothesis 3: 4 Literature Review 4 Research Design and Methodology 5 Data Analysis and Interpretations 6 Multiple Regression Analysis 6 Multiple Regression Model 6 Multiple Regression Equation 7 Relationship of R-square 8 Adjusted R-Square 8 Standard Error 8 Interpretation of Intercept b1 8 T-Stat and P-Value 9 Graphical Presentations: 9 Conclusion 13 Recommendations 13 Appendix 14 Bibliography 16 IMPACTS OF WEATHER IN ELECTRICITY CONSUMPTION IN SACRAMENTO AREA Abstract The uncertain variations in electricity consumption have insisted the policymakers to pay more attention towards its reasons in Sacramento area. The purpose of this research study is to evaluate the impact of weather conditions on electricity consumption in Sacramento area. The dependent (electricity consumption) and independent variables (Temperature, humidity, wind speed and dummy variables) were selected. Through multiple regression and graphical descriptions, the three hypotheses have been proved. A regression model has been developed that would help the policymakers to understand the relationships between the variables. Moreover, positive linear relationships have appeared between the electric consumption and the selected variables of weather conditions. Background of the Study For proper planning of future investments in energy generation sector, policy makers require forecasts of electricity demand. The demand of electricity consumption in California has more than quadrupled since 1960. There are various factors, which are contributing to the increased energy consumption in California such as uncertain regulatory environment, swift population growth and economic growth. Apart from that, climatic change is another factor, which can also influence the electric consumption thereby, affecting the long term planning of electricity supply. Since climatic changes such as lower heating demand and higher cooling demand can directly influence the energy consumption in any economy therefore, this research study will evaluate its effects on energy consumption in Sacramento, California. This research will also assist the policy makers to analyze the extent to which the weather conditions can affect the energy consumption in Sacramento. Therefore, it will be very useful to meet the challenge of energy consumption demands in Sacramento. After conducting this research study, the answers to various questions will be achieved such as, what is the impact of weather conditions on electricity consumption? What are the affects of global warming on electricity consumption? Hypotheses The research objective of this research study is to explore the following question: “What are the impacts of weather in electricity consumption in Sacramento area?” Therefore, following are the hypotheses, which have been tested in this research study: Hypothesis 1: Ho: Increasing temperature has a positive effect on electricity consumption in Sacramento area H1: Increasing temperature does not a positive effect on electricity consumption in Sacramento area Hypothesis 2: Ho: Increasing humidity has a positive effect on electricity consumption in Sacramento area H1: Increasing humidity does not a positive effect on electricity consumption in Sacramento area Hypothesis 3: Ho: Increasing wind speed has a positive effect on electricity consumption in Sacramento area H1: Increasing wind speed does not a positive effect on electricity consumption in Sacramento area Literature Review The forecasts of electricity demand in an economy are made in consideration to the factors such as technological changes, income of consumers, population growth and prices. However, researchers have found that weather conditions variables should be also used to forecast electricity demand. A huge literature work has been done to study this relationship. In California, the population growth in the hotter interior regions of the state is causing growth in cooling demand, which ultimately affects the electric power system(Sanstad and Franco).1 Hor et al. investigated the affects of weather (Temperature, wind spread, rainfall, relative humidity and sunshine), GDP and population growth on monthly electricity demand in England and Wales. Weather conditions affect the electricity industry in terms of electricity consumption and operation of the network infrastructure (Hor et al.).2 Donovan and David discovered the use of the shade trees for reducing electricity use and they demonstrated the importance of location of tree cover to reduce summertime electricity use.3 Aroonruengsawat and Maximilian also conducted a similar research and proved the impact of weather conditions through monthl billing data of consumers. Lillard and Aigner demonstrated the impact of temperature level variations on time of day (TOD) electricity consumption and air conditioning.4 They found a positive relationship between the temperature variations and TOD consumption and air conditioning. Although, the literature has supported the impact of weather conditions on electricity consumption however, to develop accurate electricity consumption forecasts in Sacramento, California there is a need to conduct this research study. Research Design and Methodology In this research study, the impact of weather conditions on electric consumption in Sacramento has been evaluated. The variables that have been selected can be categorized into categories including the dependent and independent variables. The dependent variable is the electricity consumption while the independent are climatic variables (temperature wind speed, and humidity) and social economic variables (gross domestic product and population growth). Although considering social economic variables is important because the impact of these factors cannot be ignored whilst developing the mathematic model. Since data is on hourly basis and the macroeconomic variables do not change during two months therefore, they have not been considered in the model. The formulated equation is: E = α + β1Temp + β2Humidity + β3Wind+ β4Dummy1 + β5Dummy2 + Error term In order to test the hypothesis, the historical data of all these variables has been collected. Data has been collected from published data sources. The data has been collected for two months of July and August 2008 on hourly basis. After collecting all data, regression analysis has conducted to develop a mathematical model. Based on the mathematical model the hypotheses have been tested.  Data Analysis and Interpretations The data has been analyzed through two approaches including multiple regression approach and graphical description approach. Both these approaches have played a very important role in testing the hypotheses. Multiple Regression Analysis Multiple Regression Model E = α + β1Temp + β2Humidity + β3Wind+ β4Dummy1 + Error term And E = α + β1Temp + β2Humidity + β3Wind+ β4Dummy2 + Error term Where, Variable 1 = x1 = Temperature Variable 2 = x2 = Humidity Variable 3 = x3 = Wind Variable 4 = x4 = Dummy 1, Dummy 2 E = α + β1x1 + β2x2 + β3x3 + β4d1 And E = α + β1x1 + β2x2 + β3x3 + β4d1 Multiple Regression Equation E = -2157.05 + 42.65x1 + 8.78x2 + 10.82x3 – 64.51 x4 + € (Figure 1 – Appendix) E = -2221.57 + 42.65x1 + 8.78x2 + 10.82x3 + 64.51 x4 + € (Figure 2 – Appendix) The Multiple regression equation gives some very useful predictions. The positive values of all the coefficients of the variables show that temperature, humidity and wind speed are positively related to the dependent variable Electricity. If temperature or x1 variable increases by 1 degree Fahrenheit, the electricity consumption will increase by 42.65 Megawatt/hour. If temperature increases by 2 degree Fahrenheit, the electricity consumption will increase by 42.65*2 Megawatt/hour. Similarly, if humidity increases by 1 percent, the electricity consumption will increase by 8.78 Megawatt/hour. If humidity increases by 2 percent, the electricity consumption will increase by 8.78*2 Megawatt/hour. Moreover, If wind speed or increases by 1 mph, the electricity consumption will increase by 10.82 Megawatt/hour. If wind speed increases by 2 mph, the electricity consumption will increase by 10.82*2 Megawatt/hour. The coefficient of dummy variable is -64.51 for the first dummy while it is positive for dummy 2. Therefore, from the multiple regression equation it is evident that electric consumption is positively related to temperature, humidity and wind speed. Hence, through multiple regression analysis, all of our null hypotheses have been proved. Relationship of R-square The value of R-square is 0.9081in both regression cases, which means that 90.91 percent of the variation of the Y-variable or E around E Bar (its mean) is explained by its four regressors x1, x2, x3and x4. The value of R2 shows the overall goodness of fit measures of this model. Moreover, it can be also said that the correlation between E and E-hat is 0.9529 (Multiple R). Adjusted R-Square The value of Adjusted R2 is 0.9074 and its value is always less than R2. Adjusted R2 can be calculated as follows: Adjusted R2 = R2 - (1-R2)*(k-1)/ (n-k) The difference between the value of R2 and Adjusted R2 is because of the reason that when more variables are added to the model, the value of R2 increases whereas, adjusted R2 takes into account the number of variables in the model and give the fit. Standard Error The standard error of this regression model is 138.15 and it shows the standard deviation of the term €. It can be manually calculated as squareroot (SSE/ (n-k)) and Interpretation of Intercept b1 The value of intercept is -2157 with first dummy regression and -2221 with second dummy regression. If it is theoretically explained, then it means that in the ideal case, if the values of all independent variables are zero, then energy consumption will be negative. T-Stat and P-Value The t-test describes the significance of the test and it evaluates whether the means of two groups of data are statistically different from each other or not. T-sat value of all variables is great that 2.5 which shows their significance in this regression model. Moreover, the p-value shows that how much it is possible that the coefficient of a specific independent variable has emerged by change and it is not significant to explain a real relationship. Since, the values of p for all the variables are in negative exponent therefore, it can be argued that there are extremely low chances that relationship has emerged randomly. In other words, the relationship is real. Graphical Presentations: The hypotheses have been also tested through qualitative or graphical descriptions. In order to determine the impact of each variable one by one, the graphs are as follows. Electricity versus Temperature: The Graph 1 shows the relationship between electricity consumption and temperature. It shows that between July and August 2008, the temperature was increasing with only a few exceptions. The trend line shows that the electric consumption is positively-linearly related to the temperature. A very slight increase has been noticed in the electricity consumption during these two months. Since, there is not a drastic change in temperature therefore; the electric consumption has not followed the drastic increase. However, this graph is significant enough to explain the linear positive relationship between the two variables. Therefore, the first null hypothesis is proved, saying that increasing temperature has a positive effect on electricity consumption in Sacramento area Graph 1: Electricity versus Humidity: The Graph 2 shows the relationship between electricity consumption and humidity. It shows that between July and August 2008, the humidity was increasing with only a few exceptions. The trend line shows that the electric consumption is positively-linearly related to the humidity. Again, a very slight increase has been noticed in the electricity consumption during these two months as a result of a slight increase in humidity. However, this graph is significant enough to explain the linear positive relationship between the two variables. Therefore, the second null hypothesis is proved, saying that increasing humidity has a positive effect on electricity consumption in Sacramento area. Graph 2 Electricity versus Wind: The Graph 3 shows the relationship between electricity consumption and wind speed. The trend line shows that the electric consumption is positively-linearly related to wind speed. Again, a very slight increase has been noticed in the electricity consumption during these two months as a result of a slight increase in wind speed. However, this graph is important enough to explain the linear positive relationship between the two variables. Therefore, the third null hypothesis is proved, saying that increasing wind speed has a positive effect on electricity consumption in Sacramento area. Graph 3 Moreover, the graph of dummy variables has been shown in the appendix as figure 3 and 4. Since all of the three hypotheses have been proved therefore, the impact of dummy variables is very insignificant. Conclusion Therefore, based on the data analysis and data interpretation, it can be concluded that weather conditions have a very significant impact on electric consumption in Sacramento area. The specific weather conditions including temperature, humidity and wind speed have a positive linear relationship with electric consumption. Therefore, an increase in these factors would directly increase the electricity consumption in Sacramento area while opposite would happen in other case. Recommendations Therefore, before allocating the budget for electricity generation, the policy makers should consider these factors very seriously. Since, the scope of this research study is limited because it evaluates the impact of weather conditions on electric consumption in Sacramento area during two months therefore; it is recommended that another similar study can be also conducted with a broader scope (comparing summers and winters electricity consumptions). Appendix Figure 1: Regression Analysis with Dummy 1 Figure 2: Regression Analysis with Dummy 2 Figure 3: Electricity versus Dummy1 Figure 4: Electricity versus Dummy2 Bibliography Aroonruengsawat, Anin and Maximilian Auffhammer. "IMPACTS OF CLIMATE CHANGE ON RESIDENTIAL ELECTRICITY CONSUMPTION:EVIDENCE FROM BILLING DATA." 2009. Cameron, Colin, A. "EXCEL 2007: Multiple Regression." Cameron . 9 November 2009 . Donovan, H. Geoffrey and Butry T. David. "The value of shade: Estimating the effect of urban trees on summertime electricity use." Energy and Buildings (2009): 662–668. Hor, Ching-Lai et al. "Analyzing the Impact of Weather Variables on Monthly Electricity Demand." 2005. "Interpretation in Multiple Regression." ISDS. 9 November 2009 . Lillard, A. Lee and J. Dennis Aigner. "Time-of-Day Electricity Consumption Response to Temperature and the Ownership of Air Conditioning Appliances." Journal of Business & Economic Statistics (1984): 40-53. Sanstad, H. Alan and Guido Franco. "Climate change and electricity demand in California." 2007. Read More