<![CDATA[One of the focus of the government in the 2020 macroeconomic strategy is the realization of controlled inflation, an indicator often used to measure inflation, namely the Consumer Price Index (CPI). The movement of prices of goods and services consumed by the public causes changes in the value of the CPI, when unstable price movements can cause inflation. Forecasting is used to help policy makers to be taken into consideration in order to avoid inflation instability. This research used IHK as data inputs which will be formed the pattern then did data normalization process and processed using Backpropagation Neural Network method for the forecasting of CPI, return value with data denormalization and lastly using Mean Absolute Percentage Error (MAPE) for evaluation of forecasting results. The smallest MAPE value obtained from this research is 0.463% with the value of neuron input = 6, hidden neuron value = 10, initial weight range value in the range -1 to 1, learning rate value = 0.1, and epoch value = 5000]]>
Copyrights © 2020
