SlideShare a Scribd company logo
1 of 15
Electricity Net Generation in U.S
Time series analysis and forecasting
Shen (Carol) Yan, Shih-Wen (Elsa) Huang
Motivation
 We are curious whether time series confirm to our original assumption:
winter has the highest net electricity generation.
 Dataset from EIA has 511 observations and 2 variables: month and
electricity net generation total
* EIA: Energy Information Administration
Background
 With the economic growth and industries development in the
U.S, the demand of electricity is increasing year by year. This
phenomenon leads to higher electricity generation and also
reflects on the dataset from January 1973 to July 2015:
 Increasing trend- Total of electricity net generation increase per
year.
 Seasonal behavior
39%
1%0%
27%
0%
19%
7%
7% 0%0%
2014 Electricity generation sources
coal
petroleum liquids
petroleum coke
natural gas
other gas
nuclear
hydroelectric
conventional
renewable source
pump
other
35%
1%0%
32%
0%
19%
6%
7% 0%0%
2015 Electricity generation sources (till August)
coal
petroleum liquids
petroleum coke
natural gas
other gas
nuclear
hydroelectric
conventional
renewable source
pump
other
Electricity sources
Objectives
1. The model behavior of this dataset
2. Create the fitting model to forecast the following
electricity generation in next 17 month till December
2016.
Time plot of electricity generation
 Trend: Increasing trend
 Seasonality
 Spikes - Something happened in 2009: about price
2009
Electricity net generation
decreased
Before building the model
 Detrend
 Deseaonalization
Detrend
 Detrend: Flat
 ACF & PACF:
Simultaneously show seasonality in the time period of 12 month
Deseasonalization
 ACF & PACF:
 Dickey-Fuller test:
p-value(0.01) <0.05, null hypothesis of non-stationary is rejected.
Build the model-SARIMA
 Model: ARIMA(1,1,1)(0, 1, 1)[12]
 Test of coefficients: All parameters are significant.
 Expression: (1-0.45B)(1-B)(1-B12)Xt=(1-0.90B)(1-0.73B12)
Diagnosis
 ACF plot of residuals: generally stationary
 L-jung Box tests: p-value>0.05, cannot reject White
Noise(residuals)
 Normal quantile plot:
Brief conclusion:
The model SARIMA(1,1,1)(0,1,1)[12] is statistically acceptable
and can be processed to explain and make a prediction.
Forecast
 Point forecast for following 17 months
Validation of model
 MAPE from Back-test: 1.63%
 Compare with the latest data announced by EIA
and calculate new MAPE: 0.41%
Released from EIA Our Forecast
August 2015 392298 393923
* EIA: Energy Information Administra
Fit well!
Conclusion
 This is a non-stationary model with an increasing trend.
 Model has seasonal behavior: peak period is during summer.
 The forecasts for the following 17 months are consistent with previous
patterns.
 Our model is reliable: The specific forecast of August is with minor error to
the number announced by Energy Information Administration official
website.
 Limitation:
Further research is needed on time series regression to identify impact of
each source such as, petroleum, coal, nuclear and natural gas, etc., on
electricity net generation in the U.S.
Reference
 http://www.eia.gov/todayinenergy/detail.cfm?id=8450
 http://www.eia.gov/electricity/monthly/epm_table_grapher.cfm?t=e
pmt_1_01

More Related Content

What's hot

Internship_Presentation
Internship_PresentationInternship_Presentation
Internship_PresentationSourabh Gujar
 
Lakeshore Center Heat Pump Study
Lakeshore Center Heat Pump Study Lakeshore Center Heat Pump Study
Lakeshore Center Heat Pump Study Daanish Tyrewala
 
Are the beliefs of the climate change deniers, skeptics, and trivializers sup...
Are the beliefs of the climate change deniers, skeptics, and trivializers sup...Are the beliefs of the climate change deniers, skeptics, and trivializers sup...
Are the beliefs of the climate change deniers, skeptics, and trivializers sup...Economic and Social Research Institute
 
Michal Monselise - Online change point detection using spark streaming
Michal Monselise - Online change point detection using spark streamingMichal Monselise - Online change point detection using spark streaming
Michal Monselise - Online change point detection using spark streamingPyData
 
Habitat for Humanity HVAC System Analysis
Habitat for Humanity HVAC System AnalysisHabitat for Humanity HVAC System Analysis
Habitat for Humanity HVAC System AnalysisMatt Bartholomew
 
Interference effects for inline chmineys
Interference effects for inline chmineysInterference effects for inline chmineys
Interference effects for inline chmineysFull Scale Dynamics
 
Art Data Hackathon - Klima iOS app
Art Data Hackathon - Klima iOS appArt Data Hackathon - Klima iOS app
Art Data Hackathon - Klima iOS applab_SNG
 
Energy efficiency dataset
Energy efficiency datasetEnergy efficiency dataset
Energy efficiency datasetAnkit Ghosalkar
 
What to Expect from the Paris Climate Conference
What to Expect from the Paris Climate ConferenceWhat to Expect from the Paris Climate Conference
What to Expect from the Paris Climate ConferencePCKnapp
 
Class 8 mathematical modeling of interacting and non-interacting level systems
Class 8   mathematical modeling of interacting and non-interacting level systemsClass 8   mathematical modeling of interacting and non-interacting level systems
Class 8 mathematical modeling of interacting and non-interacting level systemsManipal Institute of Technology
 
Observing Solid, Liquid and Gas Particles Day 3
Observing Solid, Liquid and Gas Particles Day 3Observing Solid, Liquid and Gas Particles Day 3
Observing Solid, Liquid and Gas Particles Day 3jmori1
 

What's hot (14)

Internship_Presentation
Internship_PresentationInternship_Presentation
Internship_Presentation
 
Lakeshore Center Heat Pump Study
Lakeshore Center Heat Pump Study Lakeshore Center Heat Pump Study
Lakeshore Center Heat Pump Study
 
Are the beliefs of the climate change deniers, skeptics, and trivializers sup...
Are the beliefs of the climate change deniers, skeptics, and trivializers sup...Are the beliefs of the climate change deniers, skeptics, and trivializers sup...
Are the beliefs of the climate change deniers, skeptics, and trivializers sup...
 
Michal Monselise - Online change point detection using spark streaming
Michal Monselise - Online change point detection using spark streamingMichal Monselise - Online change point detection using spark streaming
Michal Monselise - Online change point detection using spark streaming
 
Habitat for Humanity HVAC System Analysis
Habitat for Humanity HVAC System AnalysisHabitat for Humanity HVAC System Analysis
Habitat for Humanity HVAC System Analysis
 
Montgomery County Gude Drive Weather Station August 2017
Montgomery County Gude Drive Weather Station August 2017Montgomery County Gude Drive Weather Station August 2017
Montgomery County Gude Drive Weather Station August 2017
 
Interference effects for inline chmineys
Interference effects for inline chmineysInterference effects for inline chmineys
Interference effects for inline chmineys
 
Art Data Hackathon - Klima iOS app
Art Data Hackathon - Klima iOS appArt Data Hackathon - Klima iOS app
Art Data Hackathon - Klima iOS app
 
Energy efficiency dataset
Energy efficiency datasetEnergy efficiency dataset
Energy efficiency dataset
 
What to Expect from the Paris Climate Conference
What to Expect from the Paris Climate ConferenceWhat to Expect from the Paris Climate Conference
What to Expect from the Paris Climate Conference
 
Class 25 i, d electronic controllers
Class 25   i, d electronic controllersClass 25   i, d electronic controllers
Class 25 i, d electronic controllers
 
Class 8 mathematical modeling of interacting and non-interacting level systems
Class 8   mathematical modeling of interacting and non-interacting level systemsClass 8   mathematical modeling of interacting and non-interacting level systems
Class 8 mathematical modeling of interacting and non-interacting level systems
 
Undergraduate Modeling Workshop - Southeastern US Rainfall Working Group Fina...
Undergraduate Modeling Workshop - Southeastern US Rainfall Working Group Fina...Undergraduate Modeling Workshop - Southeastern US Rainfall Working Group Fina...
Undergraduate Modeling Workshop - Southeastern US Rainfall Working Group Fina...
 
Observing Solid, Liquid and Gas Particles Day 3
Observing Solid, Liquid and Gas Particles Day 3Observing Solid, Liquid and Gas Particles Day 3
Observing Solid, Liquid and Gas Particles Day 3
 

Similar to Electricity Net Generation

MFBLP Method Forecast for Regional Load Demand System
MFBLP Method Forecast for Regional Load Demand SystemMFBLP Method Forecast for Regional Load Demand System
MFBLP Method Forecast for Regional Load Demand SystemCSCJournals
 
Probabilistic forecasting of long-term peak electricity demand
Probabilistic forecasting of long-term peak electricity demandProbabilistic forecasting of long-term peak electricity demand
Probabilistic forecasting of long-term peak electricity demandRob Hyndman
 
Random Forest Ensemble of Support Vector Regression for Solar Power Forecasting
Random Forest Ensemble of Support Vector Regression for Solar Power ForecastingRandom Forest Ensemble of Support Vector Regression for Solar Power Forecasting
Random Forest Ensemble of Support Vector Regression for Solar Power ForecastingMohamed Abuella
 
2 session 2a_hp case study_2010_cfvg
2 session 2a_hp case study_2010_cfvg2 session 2a_hp case study_2010_cfvg
2 session 2a_hp case study_2010_cfvgkimsach
 
Time Series Assignment- Household Electricity Consumption
Time Series Assignment- Household Electricity ConsumptionTime Series Assignment- Household Electricity Consumption
Time Series Assignment- Household Electricity ConsumptionBala Gowtham
 
IMDC PRESENTATION
IMDC PRESENTATIONIMDC PRESENTATION
IMDC PRESENTATIONNauman Khan
 
Investigation effects-of-supplying-power-distrubition
Investigation effects-of-supplying-power-distrubitionInvestigation effects-of-supplying-power-distrubition
Investigation effects-of-supplying-power-distrubitionslmnsvn
 
Investigation of-effects-of-supplying-jenins-power
Investigation of-effects-of-supplying-jenins-powerInvestigation of-effects-of-supplying-jenins-power
Investigation of-effects-of-supplying-jenins-powerslmnsvn
 
Forecasting Methodology Used in Restructured Electricity Market: A Review
Forecasting Methodology Used in Restructured Electricity Market: A ReviewForecasting Methodology Used in Restructured Electricity Market: A Review
Forecasting Methodology Used in Restructured Electricity Market: A ReviewDr. Sudhir Kumar Srivastava
 
DANIEL_OWEN_POSTER
DANIEL_OWEN_POSTERDANIEL_OWEN_POSTER
DANIEL_OWEN_POSTERDaniel Owen
 
Energy Models and Scenarios - predicting Germany's electricity production sys...
Energy Models and Scenarios - predicting Germany's electricity production sys...Energy Models and Scenarios - predicting Germany's electricity production sys...
Energy Models and Scenarios - predicting Germany's electricity production sys...Justice Okoroma
 
presentation of Conference On Electrical And Electronic Engineering 2015
presentation of Conference On Electrical And Electronic Engineering 2015presentation of Conference On Electrical And Electronic Engineering 2015
presentation of Conference On Electrical And Electronic Engineering 2015sager alswed
 
Actual Penalty and Deviation Settlement Mechanism (DSM) Penalty in Interstate...
Actual Penalty and Deviation Settlement Mechanism (DSM) Penalty in Interstate...Actual Penalty and Deviation Settlement Mechanism (DSM) Penalty in Interstate...
Actual Penalty and Deviation Settlement Mechanism (DSM) Penalty in Interstate...Das A. K.
 
Machine Learning Foundations Project Presentation
Machine Learning Foundations Project PresentationMachine Learning Foundations Project Presentation
Machine Learning Foundations Project PresentationAmit J Bhattacharyya
 
Ground measured data vs meteo data sets:57 locations in India_01.01.2020
Ground measured data vs meteo data sets:57 locations in India_01.01.2020Ground measured data vs meteo data sets:57 locations in India_01.01.2020
Ground measured data vs meteo data sets:57 locations in India_01.01.2020Gensol Engineering Limited
 
Development of a solar PV energy assessment tool for EG-Audit Ltd.
Development of a solar PV energy assessment tool for EG-Audit Ltd.Development of a solar PV energy assessment tool for EG-Audit Ltd.
Development of a solar PV energy assessment tool for EG-Audit Ltd.Daniel Owen
 

Similar to Electricity Net Generation (20)

MFBLP Method Forecast for Regional Load Demand System
MFBLP Method Forecast for Regional Load Demand SystemMFBLP Method Forecast for Regional Load Demand System
MFBLP Method Forecast for Regional Load Demand System
 
Probabilistic forecasting of long-term peak electricity demand
Probabilistic forecasting of long-term peak electricity demandProbabilistic forecasting of long-term peak electricity demand
Probabilistic forecasting of long-term peak electricity demand
 
WRECON2
WRECON2WRECON2
WRECON2
 
Random Forest Ensemble of Support Vector Regression for Solar Power Forecasting
Random Forest Ensemble of Support Vector Regression for Solar Power ForecastingRandom Forest Ensemble of Support Vector Regression for Solar Power Forecasting
Random Forest Ensemble of Support Vector Regression for Solar Power Forecasting
 
2 session 2a_hp case study_2010_cfvg
2 session 2a_hp case study_2010_cfvg2 session 2a_hp case study_2010_cfvg
2 session 2a_hp case study_2010_cfvg
 
Time Series Assignment- Household Electricity Consumption
Time Series Assignment- Household Electricity ConsumptionTime Series Assignment- Household Electricity Consumption
Time Series Assignment- Household Electricity Consumption
 
IMDC PRESENTATION
IMDC PRESENTATIONIMDC PRESENTATION
IMDC PRESENTATION
 
Investigation effects-of-supplying-power-distrubition
Investigation effects-of-supplying-power-distrubitionInvestigation effects-of-supplying-power-distrubition
Investigation effects-of-supplying-power-distrubition
 
Investigation of-effects-of-supplying-jenins-power
Investigation of-effects-of-supplying-jenins-powerInvestigation of-effects-of-supplying-jenins-power
Investigation of-effects-of-supplying-jenins-power
 
Forecasting Methodology Used in Restructured Electricity Market: A Review
Forecasting Methodology Used in Restructured Electricity Market: A ReviewForecasting Methodology Used in Restructured Electricity Market: A Review
Forecasting Methodology Used in Restructured Electricity Market: A Review
 
DANIEL_OWEN_POSTER
DANIEL_OWEN_POSTERDANIEL_OWEN_POSTER
DANIEL_OWEN_POSTER
 
Energy Models and Scenarios - predicting Germany's electricity production sys...
Energy Models and Scenarios - predicting Germany's electricity production sys...Energy Models and Scenarios - predicting Germany's electricity production sys...
Energy Models and Scenarios - predicting Germany's electricity production sys...
 
presentation of Conference On Electrical And Electronic Engineering 2015
presentation of Conference On Electrical And Electronic Engineering 2015presentation of Conference On Electrical And Electronic Engineering 2015
presentation of Conference On Electrical And Electronic Engineering 2015
 
Icept 2017 abanihi
Icept 2017 abanihiIcept 2017 abanihi
Icept 2017 abanihi
 
05 2017 05-04-clear sky models g-kimball
05 2017 05-04-clear sky models g-kimball05 2017 05-04-clear sky models g-kimball
05 2017 05-04-clear sky models g-kimball
 
Actual Penalty and Deviation Settlement Mechanism (DSM) Penalty in Interstate...
Actual Penalty and Deviation Settlement Mechanism (DSM) Penalty in Interstate...Actual Penalty and Deviation Settlement Mechanism (DSM) Penalty in Interstate...
Actual Penalty and Deviation Settlement Mechanism (DSM) Penalty in Interstate...
 
Machine Learning Foundations Project Presentation
Machine Learning Foundations Project PresentationMachine Learning Foundations Project Presentation
Machine Learning Foundations Project Presentation
 
Ground measured data vs meteo data sets:57 locations in India_01.01.2020
Ground measured data vs meteo data sets:57 locations in India_01.01.2020Ground measured data vs meteo data sets:57 locations in India_01.01.2020
Ground measured data vs meteo data sets:57 locations in India_01.01.2020
 
Development of a solar PV energy assessment tool for EG-Audit Ltd.
Development of a solar PV energy assessment tool for EG-Audit Ltd.Development of a solar PV energy assessment tool for EG-Audit Ltd.
Development of a solar PV energy assessment tool for EG-Audit Ltd.
 
Technology Forum - Steve Nadel
Technology Forum - Steve NadelTechnology Forum - Steve Nadel
Technology Forum - Steve Nadel
 

Electricity Net Generation

  • 1. Electricity Net Generation in U.S Time series analysis and forecasting Shen (Carol) Yan, Shih-Wen (Elsa) Huang
  • 2. Motivation  We are curious whether time series confirm to our original assumption: winter has the highest net electricity generation.  Dataset from EIA has 511 observations and 2 variables: month and electricity net generation total * EIA: Energy Information Administration
  • 3. Background  With the economic growth and industries development in the U.S, the demand of electricity is increasing year by year. This phenomenon leads to higher electricity generation and also reflects on the dataset from January 1973 to July 2015:  Increasing trend- Total of electricity net generation increase per year.  Seasonal behavior
  • 4. 39% 1%0% 27% 0% 19% 7% 7% 0%0% 2014 Electricity generation sources coal petroleum liquids petroleum coke natural gas other gas nuclear hydroelectric conventional renewable source pump other 35% 1%0% 32% 0% 19% 6% 7% 0%0% 2015 Electricity generation sources (till August) coal petroleum liquids petroleum coke natural gas other gas nuclear hydroelectric conventional renewable source pump other Electricity sources
  • 5. Objectives 1. The model behavior of this dataset 2. Create the fitting model to forecast the following electricity generation in next 17 month till December 2016.
  • 6. Time plot of electricity generation  Trend: Increasing trend  Seasonality  Spikes - Something happened in 2009: about price 2009 Electricity net generation decreased
  • 7. Before building the model  Detrend  Deseaonalization
  • 8. Detrend  Detrend: Flat  ACF & PACF: Simultaneously show seasonality in the time period of 12 month
  • 9. Deseasonalization  ACF & PACF:  Dickey-Fuller test: p-value(0.01) <0.05, null hypothesis of non-stationary is rejected.
  • 10. Build the model-SARIMA  Model: ARIMA(1,1,1)(0, 1, 1)[12]  Test of coefficients: All parameters are significant.  Expression: (1-0.45B)(1-B)(1-B12)Xt=(1-0.90B)(1-0.73B12)
  • 11. Diagnosis  ACF plot of residuals: generally stationary  L-jung Box tests: p-value>0.05, cannot reject White Noise(residuals)  Normal quantile plot: Brief conclusion: The model SARIMA(1,1,1)(0,1,1)[12] is statistically acceptable and can be processed to explain and make a prediction.
  • 12. Forecast  Point forecast for following 17 months
  • 13. Validation of model  MAPE from Back-test: 1.63%  Compare with the latest data announced by EIA and calculate new MAPE: 0.41% Released from EIA Our Forecast August 2015 392298 393923 * EIA: Energy Information Administra Fit well!
  • 14. Conclusion  This is a non-stationary model with an increasing trend.  Model has seasonal behavior: peak period is during summer.  The forecasts for the following 17 months are consistent with previous patterns.  Our model is reliable: The specific forecast of August is with minor error to the number announced by Energy Information Administration official website.  Limitation: Further research is needed on time series regression to identify impact of each source such as, petroleum, coal, nuclear and natural gas, etc., on electricity net generation in the U.S.

Editor's Notes

  1. Life experience: electricity bill is always high during winter
  2. In the following discusion we can observe the time series has increasing trend and seasonal behavior
  3. Conduct timpe plot with ts Spikes, tend, seasonality The price of natural gas dropped in 2009 so industries tend to use natural gas to generate electricity rather than using coal So the total electricity temporary decreased in 2009, here is the spike
  4. First diff Very sig seasonality in acf
  5. After first difference and deseason, ACF decays to 0, indicating stationarity, so does df show although here are still spiky at certain lags which show very strong seasonality , the ACF value is between 0.1 and -0.1, which is very tiny, we think this is fine so we go to the next step to build the model
  6. We use auto arima bic criteria obtain this seasonal model : the arima model is (1,1,1) and the seasonal order is (0,1,1) the time period is 12 The next step we do model fitting and here is the is coeff test show are parameters are sig
  7. This normal plot can capture most of values except some extreme ones, so the residuals closed to normal distribution
  8. Lowest electricity generation is during spring
  9. Fit well!!!!!!!!!
  10. According to the ACF PACF plot, we find the TS has very large shock indicating that they will experience long time to converge to mean. So