<![CDATA[Natural gas is one of the world's energy sources whose use in the future is expected to increase and become an important energy source. Due to its large role in world energy, natural gas is considered a major industry in the energy market, however, natural gas price trends are very volatile and very difficult to analyze and predict. Linear regression is a statistical method that makes predictions using the development of mathematical relationships between variables, namely the independent variable (X) and the dependent variable (Y). This analysis is to determine the direction of the relationship between the independent variable and the dependent variable, whether positive or negative and to predict the value of the dependent variable if the value of the independent variable increases or decreases. The advantage of the linear regression method itself is that this method can carry out analysis using several independent variables so that it can be used to predict data easily. Data was collected through historical searches for natural gas through website sources. Natural gas prices from 2015 to 2024 were collected using test and learn models, each with 2460 rows of data. The Anaconda Navigator application is used to carry out the data analysis process. The results of this research show that machine learning can be used to predict and analyze natural gas prices with an MSE of 2569.1, MAE of 33,723, and R-Squared of 0.9856. Thus, these results show that the linear regression method has an accuracy level of 98.56% according to the value R-Squared. Keywords: NaturalGas, Prediction, Linear Regression]]>
Copyrights © 2024
