Inferensi
Vol 7, No 1 (2024)

Prediction of Rupiah Exchange Rate Against US Dollar Using Kernel-Based Time Series Approach

Sifa, Ghisella Asy (Statistics Study Program, Department of Mathematics, Universitas Airlangga)
Galena, Marcelena Vicky (Statistics Study Program, Department of Mathematics, Universitas Airlangga)
Mardianto, M. Fariz Fadillah (Statistics Study Program, Department of Mathematics, Universitas Airlangga)
Pusporani, Elly (Statistics Study Program, Department of Mathematics, Universitas Airlangga)



Article Info

Publish Date
31 Mar 2024

Abstract

Fluctuations in the rupiah exchange rate against the United States Dollar from 2020 to early 2024 have been analyzed using classical and modern time series approaches. In this study, the classical time series approach based on Gaussian Kernel successfully provides predictions with an RMSE value of 57.5722 and a MAPE of 0.29%. Meanwhile, the modern approach with RBF Kernel SVR shows an RMSE value of 74.9201 and a MAPE of 0.41%. The results of the model performance comparison show the superiority of the classical approach with the Gaussian Kernel in predicting the rupiah exchange rate against the US Dollar as an impact of the Federal Funds Rate (FFR) policy. Therefore, it is recommended to use the classical time series method based on the Gaussian Kernel in dealing with the impact of the FFR policy to improve the accuracy of predicting the Rupiah exchange rate against the United States Dollar. This research supports the achievement of the 8th Sustainable Development Goals (SDGs) related to economic and social matters while providing a better understanding of currency exchange rate fluctuations and providing recommendations that can help in managing economic risks related to global monetary policy.

Copyrights © 2024






Journal Info

Abbrev

inferensi

Publisher

Subject

Computer Science & IT Decision Sciences, Operations Research & Management Engineering Mathematics Social Sciences

Description

The aim of Inferensi is to publish original articles concerning statistical theories and novel applications in diverse research fields related to statistics and data science. The objective of papers should be to contribute to the understanding of the statistical methodology and/or to develop and ...