Greenhouse gas emissions (GHG) generated by human activities have negative impacts such as climate change, global warming, threats to food security, natural disasters, and rising sea levels. Several countries have implemented measures to reduce GHG emissions, including reforestation, environmentally friendly industries, and providing incentives to initiatives that reduce GHG emissions. A carbon price prediction is provided to optimize carbon incentives for investors. Carbon price fluctuation on global market conditions requires methods to predict the prices accurately. The methods utilize carbon price data from European Energy Exchange (EEX) carbon market. The data employed in this research encompasses the closing prices of carbon. This research applies two algorithms, Deep Multilayer Perceptron (DMLP) and Long Short-Term Memory (LSTM), to forecast carbon prices in time series predictions. These models use the closing price variable. The simulation results show that the Deep Multilayer Perceptron method is more accurate than the Long Short-Term Memory method. The method with higher performance will result in better risk measurement of carbon price.