ANALISIS PENGARUH INFLASI TERHADAP BI RATE di INDONESIA MENGGUNAKAN METODE VECTOR AUTOREGRESSIVE

SITORUS, SELINA (2025) ANALISIS PENGARUH INFLASI TERHADAP BI RATE di INDONESIA MENGGUNAKAN METODE VECTOR AUTOREGRESSIVE. S1 thesis, UNIVERSITAS UNJA.

[img] Text
Ringkasan.pdf

Download (247kB)
[img] Text
BAB 5.pdf

Download (252kB)
[img] Text
BAB 1.pdf

Download (274kB)
[img] Text
Daftar Pustaka.pdf

Download (149kB)
[img] Text
Lembar Pengesahan.pdf

Download (900kB)
[img] Text
COVER.pdf

Download (666kB)
[img] Text
SKRIPSI FULL_SELINA FEBIYANTI_F1C221045.pdf
Restricted to Repository staff only

Download (8MB)
Official URL: https://repository.unja.ac.id/

Abstract

Inflasi dan BI Rate merupakan dua indikator ekonomi penting yang saling memengaruhi dalam proses penentuan kebijakan moneter di Indonesia. Hubungan antara keduanya bersifat dua arah, di mana inflasi dapat mendorong penyesuaian BI Rate, dan sebaliknya. Namun dalam praktiknya, arah dan kekuatan pengaruh tersebut tidak selalu berjalan konsisten dari waktu ke waktu, bisa saja variabel inflasi sebagai variabel dependen terhadap BI Rate atau sebaliknya. Ketidakpastian dalam pola hubungan ini menjadi tantangan tersendiri karena dapat menyebabkan keterlambatan pengambilan keputusan serta menurunkan efektivitas kebijakan yang diterapkan. Selain itu, hubungan timbal balik ini berlangsung dalam kurun waktu tertentu dan menunjukkan ketergantungan antar periode, sehingga termasuk dalam fenomena deret waktu (time series) yang memerlukan pendekatan analisis yang mempertimbangkan dinamika antar variabel dalam dimensi waktu. Penelitian ini menggunakan metode Vector Autoregressive (VAR) untuk menganalisis keterkaitan antara inflasi dan BI Rate serta melakukan peramalan terhadap pergerakannya di masa mendatang. Model VAR dipilih karena dapat membentuk sistem persamaan simultan yang merepresentasikan hubungan timbal balik antar variabel endogen tanpa harus menetapkan variabel independen atau dependen secara eksplisit. Permasalahan matematis yang diangkat mencakup kestasioneran data, pemilihan lag optimal berdasarkan Akaike Information Criterion (AIC), pengujian stabilitas model melalui akar karakteristik, serta proses peramalan time series menggunakan struktur persamaan dinamis multivariat. Data yang digunakan dalam penelitian ini berupa data bulanan inflasi dan BI Rate periode Januari 2020 hingga April 2025. Model terbaik yang diperoleh adalah VAR(2). Hasil uji Kausalitas Granger menunjukkan adanya hubungan dua arah antara inflasi dan BI Rate. Impulse Response Function (IRF) memperlihatkan bahwa inflasi merespons kenaikan BI Rate dengan penurunan dalam beberapa periode, sementara BI Rate merespons shock inflasi dengan peningkatan secara bertahap. Variance Decomposition (VD) menunjukkan bahwa inflasi lebih banyak dipengaruhi oleh dirinya sendiri, namun kontribusinya terhadap variasi BI Rate meningkat dari waktu ke waktu. Akurasi peramalan model juga cukup baik, ditunjukkan oleh nilai Root Mean Square Error (RMSE) sebesar 1,459 untuk inflasi dan 0,260 untuk BI Rate. Hasil penelitian ini menunjukkan bahwa model VAR dapat menjadi alat bantu kuantitatif dalam memahami dinamika hubungan antara inflasi dan BI Rate, serta mendukung perumusan kebijakan moneter berbasis data historis. Inflation and the BI Rate are two important economic indicators that affect each other in the process of determining monetary policy in Indonesia. The relationship between the two is two-way, where inflation can prompt an adjustment of the BI Rate, and vice versa. However, in practice, the direction and strength of the influence do not always run consistently from time to time, it can be the inflation variable as a dependent variable on the BI Rate or vice versa. Uncertainty in this relationship pattern is a challenge in itself because it can cause delays in decision-making and reduce the effectiveness of the policies implemented. In addition, this reciprocal relationship lasts for a certain period of time and shows dependency between periods, so it is included in the time series phenomenon which requires an analytical approach that considers the dynamics between variables in the time dimension. This study uses the Vector Autoregressive (VAR) method to analyze the relationship between inflation and the BI Rate and forecast its future movements. The VAR model was chosen because it can form a system of simultaneous equations that represent the reciprocal relationships between endogenous variables without having to explicitly define independent or dependent variables. The mathematical problems raised include data stationary, optimal lag selection based on the Akaike Information Criterion (AIC), testing the stability of the model through characteristic roots, and the time series forecasting process using a multivariate dynamic equation structure. The data used in this study is in the form of monthly inflation data and BI Rate for the period from January 2020 to April 2025. The best model obtained is . The results of the Granger Causality test show that there is a two-way relationship between inflation and the BI Rate. VAR(2)The Impulse Response Function (IRF) shows that inflation responds to the BI Rate hike with a decrease in several periods, while the BI Rate responds to inflation shocks with a gradual increase. Variance Decomposition (VD) shows that inflation is more influenced by itself, but its contribution to the variation of the BI Rate increases over time. The model's forecasting accuracy is also quite good, as shown by the Root Mean Square Error (RMSE) value of 1.459 for inflation and 0.260 for the BI Rate. The results of this study show that the VAR model can be a quantitative tool in understanding the dynamics of the relationship between inflation and the BI Rate, as well as supporting the formulation of monetary policy based on historical data.

Type: Thesis (S1)
Uncontrolled Keywords: VAR, Inflasi, BI Rate, dan Peramalan
Subjects: L Education > L Education (General)
Divisions: Fakultas Sains dan Teknologi > Matematika
Depositing User: SELINA FEBIYANTI BR. SITORUS
Date Deposited: 15 Jul 2025 03:17
Last Modified: 06 Oct 2025 03:11
URI: https://repository.unja.ac.id/id/eprint/84147

Actions (login required)

View Item View Item