Big Data for Financial Analysis: Inflation Rate Forecasting using Random Forest

Approaching the end year of 2022, the increasing inflation rate causes the world to be in turmoil. Abnormal goods and services pricing affects consumer buying power, which disrupts the capital flows of the country. To prevent the disruption, the inflation rate must be predicted each month, in purpose of implementing financial policy before the capital flows disruption happens. This paper seeks to predict the inflation rate from goods and services data in Indonesia using Random Forest model for regression. To predict the inflation rate, this paper will discuss the data source that was gathered for prediction input, data preprocessing, split dataset to training and testing, and applying the machine learning model. The experiment result shows that random forest in the forecasting inflation rate, tuned and not tuned, show better results than GBT, proven with better RMSE at 0.681693034 (without tuning) and 0.631512634 (with tuning). But for MAE, does not show better results than GBT and GLR at 0.553645579 (without tuning) and 0.516586905 (with tuning).