Time Series Modeling and Forecasting on Carbon dioxide Emission in Sri Lanka

Abstract

The continuous rise of anthropogenic carbon dioxide (CO2) atmospheric emissions is a major cause of global warming with adverse environmental effects. Most developed countries have a higher share of annual CO2 emissions in global emissions. As a developing country, Sri Lanka’s annual CO2 emission levels are lowest, but the total annual CO2 emission has increased at an annual growth rate of 5.06%. As a signatory of the Kyoto Protocol and the Paris Agreement, and being highly vulnerable to climate change, Sri Lanka commits to reduce its CO2 emissions. Valid database analysis of CO2 emission modeling and forecasting in Sri Lanka will help to policy makers in reducing CO2 emissions in Sri Lanka. In the present study, different Autoregressive Integrated Moving Average (ARIMA) models were developed to model the CO2 emission by using time series data from 1950-2019. The performance of these developed models was assessed with different selection measure criteria, and the model having the minimum value of these criteria was considered as the best forecasting model. Based on findings, ARIMA (0, 1, 1) is the best fitted model in predicting the emission of CO2 in Sri Lanka. Vector Auto Regressive with exogenous variable (VARX) model was used to assess the impact of energy consumption, GDP and urban population on CO2 emission in Sri Lanka with the time series secondary data from 1965 to 2019. Based on the results, the VARX (1,3) model was the best model for the relationship among these variables.