Rasyadi, Izzan (2018) Model Hybrid Exponential Smoothing Pendekatan State Space dengan Metode Time Series Regression untuk Peramalan Cash Outflow PT. Taspen (Persero). Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211440000042-Undergraduate_Theses.pdf - Accepted Version Download (2MB) | Preview |
Abstract
PT. Taspen Persero pada setiap bulannya melakukan suatu proyeksi yang hasil dari proyeksi tersebut digunakan sebagai nilai kas untuk 1 bulan kedepannya. Namun, sampai saat ini belum ada standar yang digunakan untuk meramalkan nilai kas perusahaan. Penelitian ini bertujuan untuk meramalkan arus kas yang ada di PT. Taspen (Persero) Kantor Cabang Utama Surabaya menggunakan pemodelan hybrid exponential smoothing pendekatan state space dengan metode time series regression. Pada penelitian ini dilakukan 2 kajian, kajian simulasi dilakukan untuk mengetahui sejauh mana metode yang digunakan mampu menangkap berbagai pola data yang mengandung efek trend, musiman dan variasi kalender dan noise. Sedangkan kajian terapan menggunakan data riil yaitu data cash outflow pada PT. Taspen (Persero) KCU Surabaya. Pemodelan terbaik dipilih menggunakan kriteria RMSEP dan sMAPEP. Pemodelan hibrida memiliki nilai RMSEP dan sMAPEP lebih kecil yaitu sebesar 7,92 dan 8,37 serta mampu mereduksi RMSEP sebesar 33,2% dan sMAPEP sebesar 44,6% dibandingkan dengan metode exponential smoothing dan ARIMAX.Peramalan menggunakan metode hibrida exponential smoothing pendekatan state space dengan metode time series regression dipilih sebagai metode terbaik untuk meramalkan arus kas PT. Taspen (Persero).
===================================================================================================
PT. Taspen (Persero)for every each month conduct a projection where the results of the projection is used as a cash value for 1 month ahead. However, until now there has been no standard used to predict the company's cash value. This study aims to predict cash flow in PT. Taspen (Persero) Main Branch Office Surabaya using hybrid exponential smoothing state space approach with the method of time series regression model. In this research there are two studies, a simulation study was conducted to find out how far the method used was able to capture various data patterns that contain effects of calendar variation and noise. While the applied study using real data that is cash outflow data at PT. Taspen (Persero) KCU Surabaya. The best modeling is selected using RMSEP and sMAPEP criteria. Hybrid modeling has a smaller error value that is equal to 7.92 for RMSEP and 8.37 for sMAPEP also can reduce the value of RSMEP up to 33,2% and sMAPEP up to 44,6% compared with exponential smoothing and ARIMAX method.Forecasting using the hybrid method of exponential smoothing state space approach with the method of time series regression is chosen as the best method to predict cash flow of PT. Taspen (Persero).
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Uncontrolled Keywords: | ARIMAX, Exponential Smoothing, Taspen, Time Series Regression,State Space, Variasi Kalender |
| Subjects: | H Social Sciences > H Social Sciences (General) > H61.4 Forecasting in the social sciences H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA30.3 Time-series analysis H Social Sciences > HA Statistics > HA31.3 Regression. Correlation |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Izzan Rasyadi |
| Date Deposited: | 05 Aug 2021 22:19 |
| Last Modified: | 05 Aug 2021 22:19 |
| URI: | http://repository.its.ac.id/id/eprint/57094 |
Actions (login required)
 |
View Item |