Peramalan Emisi Karbon Menggunakan Metode SARIMA dan LSTM

The majority of greenhouse gas (GHG) effects are caused by very high levels carbon emissions in the world. Therefore, it is necessary to take action control In this study, world's emission were forecasted based on time series data from 1949 2018 North America. This study uses 2 forecasting methods, namely SARIMA and LSTM, with consideration that both methods considered capable providing good results. Forecasting results show best parameter for [(0,1,0) (1,1,0)12] a MAPE 1.995%. Meanwhile, if you use LSTM method parameters 1 input, 4 hidden layers, output 1, produces 0.540%. condition makes more optimal predicting