Analisis Pengaruh Variabel Ekonomi Terhadap Jumlah Penumpang Kereta Api Bima Dengan Pendekatan Autoregressive Distributed Lag (ARDL)

Briantama, Raihan Abiyyu (2026) Analisis Pengaruh Variabel Ekonomi Terhadap Jumlah Penumpang Kereta Api Bima Dengan Pendekatan Autoregressive Distributed Lag (ARDL). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003221092-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Transportasi kereta api memegang peran vital dalam mobilitas di Indonesia. Penelitian ini bertujuan memodelkan pengaruh variabel makroekonomi terhadap keseimbangan jangka panjang maupun dinamika jangka pendek pergerakan penumpang Kereta Api (KA) Bima periode 2020 hingga 2025 guna mengisi kesenjangan literatur yang saat ini didominasi pendekatan peramalan. Data berfrekuensi rendah pada Produk Domestik Bruto (PDB) dan Populasi diinterpolasi menggunakan metode Denton-Cholette dan Cubic Spline. Analisis dieksekusi mengaplikasikan metode Autoregressive Distributed Lag (ARDL) yang tangguh terhadap data berordo campuran dan patahan struktural. Secara deskriptif terdeteksi volatilitas penumpang yang ekstrem akibat krisis pandemi. Variabel Populasi dieliminasi untuk mengatasi kendala multikolinearitas struktural dengan PDB. Seluruh variabel tersisa dipastikan stasioner pada campuran ordo I(0) dan I(1) serta terbukti terkointegrasi jangka panjang melalui Bounds Test dengan nilai 18,3740 di atas batas atas 4,8068. Estimasi model terbaik ARDL(1, 0, 0) mengungkap bahwa pada ekuilibrium jangka panjang, PDB terbukti berpengaruh positif dan sangat signifikan pada taraf 1% dengan nilai elastisitas sebesar 15,5364. Angka ini mengonfirmasi layanan KA Bima sebagai barang normal kelas superior yang sangat responsif, di mana kenaikan PDB 1% akan meningkatkan penumpang sebesar 15,54%. Sementara itu, harga BBM Pertamax bernilai 0,5230 merefleksikan efek substitusi namun tidak signifikan. Pada dinamika jangka pendek (ECM), Dummy Covid-19 bernilai 1,3236 mencerminkan fase adaptasi mobilitas esensial, sedangkan Dummy Lebaran bernilai -0,4691, dengan taraf signifikan 10% versi HAC memicu fenomena delayed travel dan revenge travel (rebound effect). Sistem operasional KA Bima terbukti sangat tangguh dengan koefisien Error Correction Term sebesar -0,5469 yang mengoreksi 54,69% ketidakseimbangan per bulan menuju ekuilibrium utuh dalam waktu 1,83 bulan. Keberadaan gangguan autokorelasi dan heteroskedastisitas pada OLS awal diatasi secara sempurna menggunakan estimator Robust Newey-West HAC, serta kelolosan uji stabilitas CUSUM menggaransi keandalan seluruh parameter model.
=======================================================================================================================================
Railway transportation plays a vital role in mobility in Indonesia. This study aims to model the effect of macroeconomic variables on both the long-run equilibrium and short-run dynamics of Bima Train passenger movement for the 2020–2025 period to fill the literature gap currently dominated by forecasting approaches. Low-frequency data on Gross Domestic Product (GDP) and Population were interpolated using the Denton-Cholette and Cubic Spline methods. The analysis was executed by applying the Autoregressive Distributed Lag (ARDL) method, which is robust against mixed-order data and structural breaks. Descriptively, extreme passenger volatility was detected due to the pandemic crisis. The Population variable was eliminated to resolve structural multicollinearity issues with GDP. All remaining variables were confirmed to be stationary at a mixed order of I(0) and I(1) and were proven to be cointegrated in the long run through the Bounds Test with a value of 18.3740 above the upper bound of 4.8068. Estimation of the best ARDL(1, 0, 0) model revealed that in long-run equilibrium, GDP is proven to have a positive and highly significant effect at the 1% level with an elasticity value of 15.5364. This figure confirms Bima Train services as a superior normal good that is highly responsive, where a 1% increase in GDP increases passenger volume by 15.54%. Meanwhile, the Pertamax fuel price value of 0.5230 reflects a substitution effect, though not significant. In short-run dynamics (ECM), the Covid-19 Dummy value of 1.3236 reflects an adaptive essential mobility phase, whereas the Lebaran Dummy value of -0.4691, significant at the 10% level under HAC, triggers delayed travel and revenge travel phenomena (rebound effect). The Bima Train operational system proved to be highly resilient, with an Error Correction Term coefficient of -0.5469, correcting 54.69% of disequilibrium per month toward full equilibrium in 1.83 months. The presence of autocorrelation and heteroskedasticity issues in initial OLS was perfectly resolved using the Robust Newey-West HAC estimator, and passing the CUSUM stability test guarantees the reliability of all model parameters.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Autoregressive Distributed Lag, Deret Waktu, Jumlah Penumpang, Kereta Api Bima, Autoregressive Distributed Lag, Bima Train, Newey-West HAC, Passenger Volume, Time Series
Subjects:	H Social Sciences > HA Statistics > HA30.3 Time-series analysis
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Raihan Abiyyu Briantama
Date Deposited:	04 Aug 2026 10:35
Last Modified:	04 Aug 2026 10:35
URI:	http://repository.its.ac.id/id/eprint/143103

Actions (login required)

View Item