Paper Title
Forecasting of Currency Outflow and Inflowin Bank Indonesia Based on Two Level Arimax, FFNN, and Hybrid

Abstract
Currency is the rupiah currency issued by Bank Indonesia. As one means of payment in cash, currency still plays an important role where there are many people who choose to use currency for the purposes of economic transactions. This is reflected in the increase of money in circulation (UYD) and the flow of currency through Bank Indonesia. Bank Indonesia (BI) is the central bank of the Republic of Indonesia. BI is the only institution authorized to circulate the rupiah to the public. Rupiah currency circulation conducted by Bank Indonesia in accordance with the need of the money supply. Therefore, forecasting the amount of circulation of currency required by BI. In research conducted forecasting the circulation of currency in East Nusa Tenggara (NTT) using ARIMAX, Neural Network, and hybrid ARIMAX-NN-GARCH. Criteria for selection of the best model is calculated using RMSE and sMAPE. Hybrid Model-FFNN ARIMAX is the best model for forecasting the outflow of currency while FFNN the model is the best model for forecasting inflow of currency. Keywords: TWO LEVEL ARIMAX, FFNN, HYBRID ARIMAX-FFNN, OUTFLOW, INFLOW