J Statistika : Jurnal Ilmiah Teori Dan Aplikasi Statistika
Vol 18 No 2 (2025): Jurnal Ilmiah Teori dan Aplikasi Statistika

Forecasting Taxpayer Registration Using ARIMA Models: A Case Study of KPP Pratama Meulaboh




Article Info

Publish Date
31 Dec 2025

Abstract

The number of registered taxpayers (WP) in a given region serves as a critical indicator of both the effectiveness of tax administration and the level of public participation in fulfilling tax obligations. KPP Pratama Meulaboh, as one of the vertical units of the Directorate General of Taxes, has experienced year-to-year fluctuations in taxpayer registration, highlighting the need for a systematic analytical approach to support forward-looking administrative planning. This study aims to forecast the number of registered corporate taxpayers at KPP Pratama Meulaboh using a time series approach based on the Autoregressive Integrated Moving Average (ARIMA) model. The dataset consists of annual observations covering the period 1982–2024. The analysis followed a structured procedure, including variance-stabilizing transformation, stationarity testing using the Augmented Dickey–Fuller (ADF) test, model identification through autocorrelation and partial autocorrelation analysis, and model selection based on the Akaike Information Criterion (AIC). Several ARIMA specifications were evaluated, and ARIMA(1,0,1) was selected as the optimal model, yielding the lowest AIC value (–66.0104) and statistically significant parameters. Diagnostic checks confirmed that the model residuals satisfied the white noise assumption. The selected ARIMA(1,0,1) model was subsequently used to generate ten-year-ahead forecasts, which indicate a steady upward trend in the number of registered corporate taxpayers over the forecast horizon. These results provide practical insights for tax authorities in planning administrative capacity, strengthening compliance monitoring, and supporting strategic tax base expansion within the jurisdiction of KPP Pratama Meulaboh.

Copyrights © 2025






Journal Info

Abbrev

jstatistika

Publisher

Subject

Decision Sciences, Operations Research & Management Economics, Econometrics & Finance

Description

This journal publishes scientific articles in the form of research results, case studies, or literature reviews on various aspects related to the field of statistics, scientific data and their applications. Such as Computing, Time Series, Multivariate, Data Mining, Biostatistics, Survival Analysis, ...