PERBANDINGAN MODEL AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (ARIMA) DAN VECTOR AUTOREGRESSIVE (VAR) DALAM PERAMALAN PENJUALAN TINTA PRINTER ISI ULANG

FARADILLAH TSALITS, . (2026) PERBANDINGAN MODEL AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (ARIMA) DAN VECTOR AUTOREGRESSIVE (VAR) DALAM PERAMALAN PENJUALAN TINTA PRINTER ISI ULANG. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text COVER.pdf Download (1MB)
	Text BAB I.pdf Download (234kB)
	Text BAB II.pdf Restricted to Registered users only Download (672kB) | Request a copy
	Text BAB III.pdf Restricted to Registered users only Download (280kB) | Request a copy
	Text BAB IV.pdf Restricted to Registered users only Download (484kB) | Request a copy
	Text BAB V.pdf Restricted to Registered users only Download (219kB) | Request a copy
	Text DAFTAR PUSTAKA.pdf Download (241kB)
	Text LAMPIRAN.pdf Restricted to Registered users only Download (641kB) | Request a copy

Abstract

Peramalan penjualan yang akurat diperlukan untuk meningkatkan efisiensi produksi dan menyesuaikan jumlah produksi dengan permintaan pasar agar meminimalkan risiko kelebihan stok. Penelitian ini menerapkan dua pendekatan pemodelan statistik, yaitu model univariat Autoregressive Integrated Moving Average (ARIMA) dan model multivariat Vector Autoregressive (VAR) dalam peramalan penjualan tinta printer isi ulang. Tujuan penelitian ini untuk mengetahui model terbaik untuk meramalkan penjualan tinta printer isi ulang jenis warna black dan color dari PT XYZ Refill Printer Ink. Data bulanan periode Januari 2018 hingga Desember 2025 dibagi menjadi 84 data training dan 12 data testing. Tahapan penelitian dilakukan secara sistematis mulai dari transformasi Box-Cox, uji ADF, differencing, identifikasi ACF dan PACF, identifikasi model ARIMA maupun VAR, estimasi parameter, uji signifikansi parameter, uji white noise, uji Jarque-Bera, uji kausalitas Granger, pemilihan model terbaik dan peramalan pada model ARIMA dan VAR, serta perhitungan akurasi peramalan dengan MAPE dan RMSE. Pemrosesan hingga analisis data menggunakan software Minitab dan EViews. Hasil penelitian menunjukkan bahwa model VAR memberikan performa yang lebih baik dibandingkan model ARIMA untuk peramalan 12 periode ke depan. Model VAR(3) menghasilkan nilai MAPE sebesar 6,20% untuk variabel tinta black dan 5,02% untuk variabel tinta color, sedangkan ARIMA(1,1,1) menghasilkan nilai MAPE sebesar 17,65% untuk variabel tinta black dan 18,76% untuk variabel tinta color. Hal yang sama juga pada perhitungan nilai RMSE, variabel tinta black pada model VAR(3) sebesar 23,54 yang lebih rendah daripada model ARIMA(1,1,1) sebesar 75,60, begitu pula untuk variabel tinta color pada model VAR(3) sebesar 12,95 yang lebih rendah daripada model ARIMA(1,1,1) sebesar 37,75. ***** Accurate sales forecasting is essential to enhance production efficiency and align production volume with market demand to minimize the risk of overstocking. This research applies two statistical modeling approaches, namely the univariate Autoregressive Integrated Moving Average (ARIMA) model and the multivariate Vector Autoregressive (VAR) model in forecasting refill printer ink sales. The purpose of this research is to determine the best model for predicting sales of black and color refill printer ink from PT XYZ Refill Printer Ink. Monthly data from the period of January 2018 to December 2025 are divided into 84 training data and 12 testing data. The research stages were conducted systematically, starting from the Box-Cox transformation, ADF test, differencing, identification of ACF and PACF, identification of the ARIMA and VAR models, parameter estimation, parameter significance testing, White Noise test, Jarque-Bera test, Granger causality test, selection of the best model and forecasting for both ARIMA and VAR, as well as the calculation of forecasting accuracy using MAPE and RMSE. Data processing and analysis were performed using Minitab and EViews software. The results indicate that the VAR model provides better performance than the ARIMA model for forecasting the next 12 periods. The VAR(3) model yields a MAPE of 6.20% for the black ink variable and 5.02% for the color ink variable, whereas the ARIMA(1,1,1) model yields a MAPE of 17.65% for the black ink variable and 18.76% for the color ink variable. Similarly, regarding the RMSE calculation, the black ink variable for the VAR(3) model is 23.54, which is lower than the 75.60 of the ARIMA(1,1,1) model, while for the color ink variable, the VAR(3) model yields 12.95, which is lower than the 37.75 of the ARIMA(1,1,1) model.

Item Type:	Thesis (Sarjana)
Additional Information:	1). Prof. Dr. Dian Handayani, M.Si. ; 2). Devi Eka Wardani Meganingtyas., S.Pd., M.Si.
Subjects:	Sains > Matematika Sains > Statistika
Divisions:	FMIPA > S1 Matematika
Depositing User:	Faradillah Tsalits .
Date Deposited:	24 Aug 2026 01:11
Last Modified:	24 Aug 2026 01:11
URI:	http://repository.unj.ac.id/id/eprint/72989

Actions (login required)

View Item