Medium Term Power Load Forecasting for Java and Bali Power System Using Artificial Neural Network and SARIMAX

Hutama, Andri Hardono; Akbar, Saiful; Candra, Muhammad Zuhri Catur

doi:10.1109/icodse.2018.8705837

Cited by 4 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Pada beberapa kasus, data penjualan, konsumsi listrik, jumlah wisatawan, dan sebagainya, seringkali memiliki pola tertentu, yaitu meningkat atau menurun pada waktu-waktu tertentu [12]. Hal ini mengindikasikan bahwa data tersebut memiliki pola musiman.…”

Section: Model Arima Dan Arimaxunclassified

“…Artinya, transaksi pembayaran non-tunai terbukti secara statistik dipengaruhi pula oleh transaksi pada periode sebelumnya serta variabel yang tidak dilibatkan dalam model (galat) pada periode sebelumnya. Model yang terbukti lebih sederhana dan tepat berdasarkan AIC adalah ARIMA(0,1,2)(1,0,0) 12 . Persamaan (11) adalah ARIMA(0,1,2)(1,0,0) 12 berdasarkan data training untuk transaksi pembayaran non-tunai pada waktu ke-t. Untuk menghasilkan ramalan dari model ARIMAX, maka galat pada persamaan (9) dan (10) dimodelkan terlebih dulu dengan ARIMA, lalu ditambahkan dengan estimasi dari model regresi deret waktu.…”

Section: Model Regresi Deret Waktuunclassified

“…significance limits for the partial autocorrelations) Muhammad Luthfi Setiarno Putera ARIMA(0,1,2)(1,0,0)12 …”

mentioning

confidence: 99%

See 2 more Smart Citations

Non-cash Payment Transaction Projection Using ARIMAX : Efect of Calendar

Putera

2020

JMSK

View full text Add to dashboard Cite

As the most Moslem country, economic activity in Indonesia is often parallel with the movement of Qamariah (lunar) calendar which is different with Gregorian calendar. Using calender variation, this research attempts to look for modified time series model for non-cash payment projection (forecast) aim. The result shows that calendar variation plays statistically significant role on non-cash payment, evidenced by significant payment in the month in which Eid Fitr occurs. The occurrence of Eid Fitr in the first and second week of the month is evidently characterized by increasing non-cash payment in one month earlier. The best model with highest accuracy for non-cash payment projection is ARIMAX(2,1,1) as it is able to capture the pattern, trend and fluctuation. It also suggests the peak of non-cash payment will be in December.

show abstract

Section: Model Arima Dan Arimaxunclassified

Section: Model Regresi Deret Waktuunclassified