GSTAR Modeling for Forecasting Traffic Accident Claim Frequency as a Data-Driven Policy Instrument in Malang Raya
PANGRIPTA · 14.03.2026 · 9 (1) · 54-64
DOI: 10.58411/0zjx5281 ↗Kurz gefasst
Fachartikel in der Zeitschrift PANGRIPTA, online erschienen am 14.03.2026. Verfasst von Auliya Hasna Rahadatul Aisy und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Traffic accidents represent one of the most persistent challenges of urban development, imposing a significant burden on public finances particularly through the compensation claim mechanism administered by PT Jasa Raharja. Understanding the spatial and temporal patterns of traffic accident claim frequency is a crucial foundation for local governments seeking to design effective data-driven policies. This study develops a forecasting model for traffic accident claim frequency across the Malang Raya area covering the Batu, Malang City, and Malang police jurisdictions using the Generalized Space-Time Autoregressive (GSTAR) approach as an analytical tool to support regional development planning. Monthly secondary data from PT Jasa Raharja Malang Branch covering January 2020 to March 2025 were used as the basis for modeling. The analysis proceeded through a series of systematic stages, including stationarity testing, spatial correlation and Cross Correlation Function (CCF) analysis, model identification, CCF-based spatial weight matrix construction, parameter estimation via the Seemingly Unrelated Regression (SUR) method, and residual diagnostic testing. Results indicate that the GSTAR(2,1) model successfully captures the spatio-temporal dependence structure across the three study regions, with the strongest spatial relationship identified between Batu and Malang City. Forecast results indicate that Malang consistently records the highest claim frequency, ranging from 70 to 85 claims per month. These findings carry strategic implications for local governments, the Department of Transportation, and other relevant stakeholders in optimizing resource allocation and designing more targeted and evidence-based traffic policy interventions.
Zitieren
APA
Aisy, A. H. R., Rahardjo, S., & Suprianto (2026). GSTAR Modeling for Forecasting Traffic Accident Claim Frequency as a Data-Driven Policy Instrument in Malang Raya. PANGRIPTA, 9(1), 54-64. https://doi.org/10.58411/0zjx5281
Vancouver
Aisy AHR, Rahardjo S, Suprianto. GSTAR Modeling for Forecasting Traffic Accident Claim Frequency as a Data-Driven Policy Instrument in Malang Raya. PANGRIPTA. 2026;9(1):54-64. doi:10.58411/0zjx5281
BibTeX
@article{Aisy2026,
author = {Aisy, Auliya Hasna Rahadatul and Rahardjo, Swasono and Suprianto},
title = {GSTAR Modeling for Forecasting Traffic Accident Claim Frequency as a Data-Driven Policy Instrument in Malang Raya},
journal = {PANGRIPTA},
year = {2026},
volume = {9},
number = {1},
pages = {54--64},
doi = {10.58411/0zjx5281}
}
EUVIMED