Comparison Of Support Vector Regression And Autoregressive Integrated Moving Average With Exogenous Variable On Indonesia Consumer Price Index

CPI is one of the most frequently used indicators to measure inflation rate in a region. The government can maintain economic stability by knowing value advance. Therefore, we need suitable method predict an accurate value. In this research, investigate prediction based on machine learning method, SVR, and compare it ARIMAX method. We use Indonesia data from January 2015 October 2021. SVR using four kernel functions: Radial Basis Function (RBF), Polynomial, Linear, Sigmoid. build model through auto ARIMA process. divide into two parts with three scenarios performance methods: training testing. results show that partition 80% 20% testing gives best performance. performs linear kernel, RMSE 0,743 MAPE 0,684%. (0,2,1) 1,928071 1,731598 %. From plot MAPE, predicts better than previous one-month (MA1) being influential variable next