Pemdelan Fuzzy untuk Data Time Series Menggunakan Metode Tabel Look Up dengan Transformasi Logaritma dan Diferensi dan Aplikasinya pada Data Indeks Harga Saham Gabungan (IHSG)

Abstract

The purpose of this study is to form a fuzzy model for data time series using a lookup table with
logarithmic transformation and differentiation as well as its application to predict IHSG listed on the
BSE. First, all data of IHSG is made in logarithm and differentiation. Then, the formation of the fuzzy
rules is by lookup table. In this study, the prediction of IHSG is only based on time series data of IHSG
with 144 data as the training data. The results of this study are the prediction of IHSG is based on 6 fuzzy
models formed to conclude that the model of fuzzy lookup table 8 input with logarithmic transformation is
the best model to predict IHSG. It can be seen from MAPE produced by the model is 4.09%. When it is
compared to fuzzy model 8 inputs without transformation in preceding studies, fuzzy models 8 inputs with
logarithmic transformation is still a better model because it has a smaller MAPE values.