Peramalan inflow dan outflow dengan metode regresi time series dan ARIMA untuk optimalisasi peredaran uang di Bank Indonesia Cabang Malang

Putri, Aninda Prisa Riestian (2015) Peramalan inflow dan outflow dengan metode regresi time series dan ARIMA untuk optimalisasi peredaran uang di Bank Indonesia Cabang Malang. Diploma thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
1312030061-Undergraduate.pdf

Download (7MB)

Abstract

Penelitian ini mengangkat permasalahan Bank Indonesia mengenai kondisi ekonomi yang penuh ketidakstabilan dan ketidakpastian yang cenderung mendorong agen ekonomi untuk memegang uang kartal guna berjaga-jaga dalam jumlah uang yang lebih besar. Sehingga perlunya peramalan yang baik untuk mendapatkan jumlah inflow dan outflow uang kartal yang beredar di Bank Indonesia khususnya cabang Malang. Metode yang digunakan dalam penelitian ini adalah Regresi Time Series dan ARIMA. Berdasarkan kedua metode tersebut dipilih model terbaik berdasarkan kriteria kebaikan model menggunakan RMSE. Dari hasil penelitian terlihat bahwa inflow mengalami kenaikan pada hari Selasa, Rabu, dan minggu kedua. Sedangkan outflow mengalami kenaikan pada hari Jumat dan minggu keempat. Model Regresi Time Series menunjukkan bahwa data inflow dan outflow dipengaruhi oleh dummy hari, minggu, bulan, dan dummy lebaran. Pengaruh efek Idul Fitri 10 hari karja untuk inflow setelah lebaran dan untuk outflow sebelum lebaran. Model ARIMA untuk inflow dan outflow hanya dapat menangkap seasonal hari saja dan tidak dapat menangkap efek Idul Fitri. Metode terbaik yang terpilih untuk meramalkan inflow dan outflow tahun 2015 adalah metode Regresi Time Series model 1. =========================================================================================================== The research raised the problem of Bank Indonesia about economic situation unstability and uncertainty that tend encourage agent of economic to hold money with the larger amount of money. So, being needed forecasting that is good to get the number of inflow and outflow money currency at Bank Indonesia especially Malang. In this research, methods were used is Time Series Regression and ARIMA. Based on those methods above that were choosen the best of model and based on good criteria model used RMSE. From the result of research that inflow experienced increase on Tuesday, Wednesday and second weeks while outflow experienced increase on Friday and fourth weeks. Time Series Regression showed that the data of inflow and outflow were influenced by dummy day, week, month and celebration day. The influence of celebration day is to inflow 10 work days or twice a weeks after celebration day and outflow was also influenced 10 work days or twice a weeks after celebration day. Arima model showed to inflow and outflow that only catch seasonal a day and can not catch of effect celebration day. The best method was choosen to forecasting inflow and outflow in 2015 year is time series regression model I.

Item Type: Thesis (Diploma)
Additional Information: RSSt 519.536 Put p
Uncontrolled Keywords: ARIMA; Regresi; Inflow; Outflow
Subjects: Q Science > QA Mathematics > QA278.2 Regression Analysis
Divisions: Faculty of Mathematics and Science > Statistics > (D3) Diploma
Depositing User: - Taufiq Rahmanu
Date Deposited: 24 Jun 2019 08:35
Last Modified: 24 Jun 2019 08:35
URI: http://repository.its.ac.id/id/eprint/63212

Actions (login required)

View Item View Item