Prediction Models for Carbon Dioxide Emissions and the二氧化碳排放量预测模型.pptVIP

Prediction Models for Carbon Dioxide Emissions and the Atmosphere By Chris P. Tsokos Abstract The object of the present study is to develop statistical models for predicting the carbon dioxide emissions and the atmosphere in the United States. We used monthly emissions data from 1981 to 2003 that was collected by the Carbon Dioxide Information Analysis Center. For the carbon dioxide in the atmosphere, we used the data that was collected in Mauna Loa from 1965 to 2004 by the Scripps Institution of Oceanography. The developed statistical models take into consideration trends and seasonal effects. The quality of the prediction process is illustrated using the actual data. Outline The Data The Multiplicative ARIMA Model Brief Summary of Our Procedure Evaluation Criteria Evaluation Conclusion References The CO2 Emission Data By Carbon Dioxide Information Analysis Center (CDIAC) Time: 1981 to 2003 For detailed information, see (United States Environmental Protection Agency (EPA), 2004; Marland et al., 2003) Time Series Plot on Monthly CO2 Emissions 1981-2003 The Atmospheric CO2 Data Carbon Dioxide Research Group, Scripps Institution of Oceanography, University of California Time: 1958 to 2004 Location: Mauna Loa Geographical Location of Mauna Loa Time Series Plot for Monthly CO2 in the Atmosphere 1965-2004 The Multiplicative ARIMA Model ARIMA is defined by where Brief Summary of Our Procedure on Developing the Subject Model-1 Determine the seasonal period s Check for stationarity by determining the order of differencing d, where d = 0,1,2,… according to KPSS test, until we achieve stationarity Deciding the order of the process, for our case, we let m = 5, where p + q + P + Q = m After (d, m) being selected, listing all possible configurations of (p, q, P, Q) for p + q + P + Q m Brief Summary of Our Procedure on Developing the Subject Model-2 For each set of (p, q, P, Q), estimates the parameters for each model Compute the AIC for each model, and c